In [1]:
from IPython.core.display import display, HTML
display(HTML("<style>.container { width:100% !important; }</style>"))

/tmp/ipykernel_121051/3777615979.py:1: DeprecationWarning: Importing display from IPython.core.display is deprecated since IPython 7.14, please import from IPython display
  from IPython.core.display import display, HTML

Import Libraries¶

In [2]:
import pandas as  pd
import numpy as np
import plotly
import matplotlib.pyplot as plt
import sklearn
import plotly.graph_objects as go
import plotly.express as px
pd.options.display.max_columns = 100
import seaborn as sns

from sklearn.preprocessing import LabelEncoder, StandardScaler
from sklearn.ensemble import RandomForestClassifier
from xgboost import XGBClassifier
from sklearn.metrics import  mean_absolute_error
from sklearn.model_selection import train_test_split, KFold, cross_val_score
from datetime import datetime
from catboost import Pool, CatBoostClassifier, cv
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split
In [3]:
import warnings
warnings.filterwarnings("ignore")

Import DataSet¶

In [4]:
org_data = pd.read_csv('sample_data.csv', sep=';')
In [5]:
## Copy the original data to another dataframe which can be modified later
preprocess_data = org_data.copy()

Preprocessing¶

In [9]:
preprocess_data
Out[9]:
age job marital education default balance housing loan contact day month duration campaign pdays previous poutcome y
0 58 management married tertiary no 2143 yes no unknown 5 may 261 1 -1 0 unknown no
1 44 technician single secondary no 29 yes no unknown 5 may 151 1 -1 0 unknown no
2 33 entrepreneur married secondary no 2 yes yes unknown 5 may 76 1 -1 0 unknown no
3 47 blue-collar married unknown no 1506 yes no unknown 5 may 92 1 -1 0 unknown no
4 33 unknown single unknown no 1 no no unknown 5 may 198 1 -1 0 unknown no
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
45206 51 technician married tertiary no 825 no no cellular 17 nov 977 3 -1 0 unknown yes
45207 71 retired divorced primary no 1729 no no cellular 17 nov 456 2 -1 0 unknown yes
45208 72 retired married secondary no 5715 no no cellular 17 nov 1127 5 184 3 success yes
45209 57 blue-collar married secondary no 668 no no telephone 17 nov 508 4 -1 0 unknown no
45210 37 entrepreneur married secondary no 2971 no no cellular 17 nov 361 2 188 11 other no

45211 rows × 17 columns

The dataset contains total 45,211 observations and 17 features (16 independent and 1 dependent)

Check for null values¶
In [10]:
preprocess_data.isnull().sum()
Out[10]:
age          0
job          0
marital      0
education    0
default      0
balance      0
housing      0
loan         0
contact      0
day          0
month        0
duration     0
campaign     0
pdays        0
previous     0
poutcome     0
y            0
dtype: int64

No null values are present in any of the feature.

Check for outliers¶

Age¶
In [12]:
# fig = fig
fig = px.box(preprocess_data, y="age", width=400, height = 500)
fig.show()
In [13]:
## Check the number of outiers
age_out = preprocess_data[preprocess_data['age']>70].shape[0]
age_out_perc = round((age_out/preprocess_data.shape[0])*100,2)
print("Total {} are outliers which is around {}% of total observations. \nSince the % is quite low, we can drop these observations from our analysis".format(age_out, age_out_perc))

preprocess_data  = preprocess_data[preprocess_data['previous']<=70]

Total 487 are outliers which is around 1.08% of total observations. 
Since the % is quite low, we can drop these observations from our analysis
Balance¶
In [14]:
preprocess_data.balance.value_counts(dropna=False)
Out[14]:
 0        3514
 1         195
 2         156
 4         139
 3         134
          ... 
-381         1
 4617        1
 20584       1
 4358        1
 16353       1
Name: balance, Length: 7168, dtype: int64
In [15]:
fig = fig
fig = px.box(preprocess_data, y="balance", width=400, height = 500)
fig.show()
In [16]:
balance_out = preprocess_data[preprocess_data['balance']>3462].shape[0]
balance_out_perc = round((balance_out/preprocess_data.shape[0])*100,2)
balance_out, balance_out_perc
Out[16]:
(4712, 10.42)
In [17]:
## Replace the outlier values with the mean of the columns
mean_balance = preprocess_data['balance'].mean()

preprocess_data.loc[(preprocess_data['balance']>3462), 'balance'] = mean_balance
preprocess_data.loc[(preprocess_data['balance']<-1854), 'balance'] = mean_balance
Duration¶
In [18]:
preprocess_data.duration.value_counts(dropna=False).sort_index()
Out[18]:
0        3
1        2
2        3
3        4
4       15
        ..
3366     1
3422     1
3785     1
3881     1
4918     1
Name: duration, Length: 1573, dtype: int64
In [19]:
## Duration
fig = fig
fig = px.box(preprocess_data, y="duration", width=400, height = 500)
fig.show()
In [20]:
duration_out = preprocess_data[preprocess_data['duration']>643].shape[0]
duration_out_perc = round((duration_out/preprocess_data.shape[0])*100,2)
duration_out, duration_out_perc
Out[20]:
(3235, 7.16)
In [21]:
## Replace the outlier values with the mean of the columns
mean_duration = preprocess_data['duration'].mean()

preprocess_data.loc[(preprocess_data['duration']>643), 'duration'] = mean_duration
Campaign¶
In [22]:
preprocess_data.campaign.value_counts(dropna=False).sort_index()
Out[22]:
1     17544
2     12504
3      5521
4      3522
5      1764
6      1291
7       735
8       540
9       327
10      266
11      201
12      155
13      133
14       93
15       84
16       79
17       69
18       51
19       44
20       43
21       35
22       23
23       22
24       20
25       22
26       13
27       10
28       16
29       16
30        8
31       12
32        9
33        6
34        5
35        4
36        4
37        2
38        3
39        1
41        2
43        3
44        1
46        1
50        2
51        1
55        1
58        1
63        1
Name: campaign, dtype: int64
In [23]:
## Campaign
fig = fig
fig = px.box(preprocess_data, y="campaign", width=400, height = 500)
fig.show()
In [24]:
campaign_out = preprocess_data[preprocess_data['campaign']>6].shape[0]
campaign_out_perc = round((campaign_out/preprocess_data.shape[0])*100,2)
campaign_out, campaign_out_perc
Out[24]:
(3064, 6.78)
In [25]:
## Replace the outlier values with the mean of the columns
mean_campaign = preprocess_data['campaign'].mean()

preprocess_data.loc[(preprocess_data['campaign']>6), 'campaign'] = mean_campaign
Previous¶
In [26]:
preprocess_data.previous.value_counts()
Out[26]:
0     36954
1      2772
2      2106
3      1142
4       714
5       459
6       277
7       205
8       129
9        92
10       67
11       65
12       44
13       38
15       20
14       19
17       15
16       13
19       11
20        8
23        8
18        6
22        6
24        5
27        5
21        4
29        4
25        4
30        3
38        2
37        2
26        2
28        2
51        1
58        1
32        1
40        1
55        1
35        1
41        1
Name: previous, dtype: int64
In [27]:
fig = fig
fig = px.box(preprocess_data, y="previous", width=400, height = 500)
fig.show()
In [28]:
## Check the number of outiers
previous_out = preprocess_data[preprocess_data['previous']>58].shape[0]
previous_out_perc = round((previous_out/preprocess_data.shape[0])*100,4)
print("Total {} are outliers which is around {}% of total observations. \nSince the % is quite low, we can drop these observations from our analysis".format(previous_out, previous_out_perc))

preprocess_data  = preprocess_data[preprocess_data['previous']<=58]

Total 0 are outliers which is around 0.0% of total observations. 
Since the % is quite low, we can drop these observations from our analysis
In [ ]:

Visualization¶

Date¶
In [29]:
temp1 = preprocess_data.day.value_counts().sort_index().reset_index(name = 'Count').rename(columns={'index':'Date'})
temp1
Out[29]:
Date Count
0 1 322
1 2 1292
2 3 1079
3 4 1445
4 5 1910
5 6 1932
6 7 1817
7 8 1842
8 9 1561
9 10 524
10 11 1479
11 12 1603
12 13 1585
13 14 1848
14 15 1703
15 16 1415
16 17 1939
17 18 2308
18 19 1757
19 20 2752
20 21 2026
21 22 905
22 23 939
23 24 447
24 25 840
25 26 1035
26 27 1121
27 28 1830
28 29 1745
29 30 1566
30 31 643
In [30]:
## line plot on which day most number of calls are made
import plotly.express as px
fig = px.line(temp1, x='Date', y='Count', markers=True,width=1000, height = 400,)
fig.update_layout(
   xaxis = dict(
      tickmode = 'linear',
      tick0 = 0,
      dtick = 1
   )
)

fig.show()

The company is making more number of calls majorly on 20th date in the given dataset.

Month Name¶
In [31]:
preprocess_data.month = preprocess_data.month.str.upper()
d = {'JAN':1, 'FEB':2, 'MAR':3, 'APR':4, 'MAY':5, 'JUN':6,'JUL':7, 'AUG':8, 'SEP':9, 'OCT':10, 'NOV':11, 'DEC':12}

preprocess_data['month_number'] = preprocess_data.month.map(d)

temp2 = preprocess_data.groupby(['month','month_number']).size().reset_index(name='Count').sort_values('month_number')
temp2
Out[31]:
month month_number Count
4 JAN 1 1403
3 FEB 2 2648
7 MAR 3 477
0 APR 4 2932
8 MAY 5 13766
6 JUN 6 5341
5 JUL 7 6895
1 AUG 8 6247
11 SEP 9 579
10 OCT 10 738
9 NOV 11 3970
2 DEC 12 214
In [32]:
## line plot on which day most number of calls are made
import plotly.express as px
fig = px.line(temp2, x='month', y='Count', markers=True,width=1000, height = 400,)
fig.update_layout(
   xaxis = dict(
      tickmode = 'linear',
      tick0 = 0,
      dtick = 1
   )
)

fig.show()

May has the highest number of calls, followed by July.

Correlation with Target Variables¶

Job¶
In [33]:
preprocess_data.shape
Out[33]:
(45210, 18)
In [34]:
temp1 = preprocess_data.groupby('job').size().reset_index(name='Total')
temp2 = preprocess_data.groupby(['job','y']).size().reset_index(name = 'Count').rename(columns={'y':'Subscribe'})

temp3 = temp1.merge(temp2, on=['job'])
temp3['Percentage'] = round(temp3['Count']/temp3['Total']*100,2)


fig = px.bar(temp3, x="job", y="Percentage", color='Subscribe', barmode='group',width = 1200,
             height=400, color_discrete_sequence=["#FC6955",'rgb(102,166,30)',], text = 'Percentage')
fig.update_traces(texttemplate='%{text:.2s}', textposition='outside')
fig.show()

The above plot contains the percentages of subscribers from each category of job. We can observe that, out of all calls made to each category, students have the highest percentage of subscription depending upon the calls made to student category.

Education¶

Relation of feature Education with Target Feature.

In [35]:
temp1 = preprocess_data.groupby('education').size().reset_index(name='Total')
temp2 = preprocess_data.groupby(['education','y']).size().reset_index(name = 'Count').rename(columns={'y':'Subscribe'})

temp3 = temp1.merge(temp2, on=['education'])
temp3['Percentage'] = round(temp3['Count']/temp3['Total']*100,2)


fig = px.bar(temp3, x="education", y="Percentage", color='Subscribe', barmode='group',width = 800,
             height=400, color_discrete_sequence=["#FC6955",'rgb(102,166,30)',], text = 'Percentage')
fig.update_traces(texttemplate='%{text:.2s}', textposition='outside')
fig.show()

As we can observe from the above bar plot, out of total calls made to customer having an education level of tertiary, only 15% (which is highest among other categories) subscribe to the term deposite.

Marital Status¶
In [36]:
temp1 = preprocess_data.groupby('marital').size().reset_index(name='Total')
temp2 = preprocess_data.groupby(['marital','y']).size().reset_index(name = 'Count').rename(columns={'y':'Subscribe'})

temp3 = temp1.merge(temp2, on=['marital'])
temp3['Percentage'] = round(temp3['Count']/temp3['Total']*100,2)


fig = px.bar(temp3, x="marital", y="Percentage", color='Subscribe', barmode='group',width = 800,
             height=450, color_discrete_sequence=["#FC6955",'rgb(102,166,30)',], text = 'Percentage')
fig.update_traces(texttemplate='%{text:.2s}', textposition='outside')
fig.show()

Around 15% of the singles subscribe to the term deposit over call. While divorced and married stand at 2nd and 3rd with the numbers 12% and 10%.

Month¶
In [37]:
temp1 = preprocess_data.groupby('month').size().reset_index(name='Total')
temp2 = preprocess_data.groupby(['month','y']).size().reset_index(name = 'Count').rename(columns={'y':'Subscribe'})

temp3 = temp1.merge(temp2, on=['month'])
temp3['Percentage'] = round(temp3['Count']/temp3['Total']*100,2)

temp3['month_number'] = temp3.month.map(d)
temp3 = temp3.sort_values('month_number')

fig = px.bar(temp3, x="month", y="Percentage", color='Subscribe', barmode='group',
             height=500, color_discrete_sequence=["#FC6955",'rgb(102,166,30)',], text = 'Percentage')
fig.update_traces(texttemplate='%{text:.2s}', textposition='outside')
fig.show()

March is having the highest percentage of acceptence calls, followed by December and September with percentages 47 and 46 respectively.

Correlation Plot¶

In [38]:
def plot_correlation_map( df ):
    corr = df.corr()
    _ , ax = plt.subplots( figsize =( 12,8 ) )
    cmap = sns.diverging_palette( 220 , 10 , as_cmap = True )
    _ = sns.heatmap(
        corr, 
        cmap = cmap,
        square=True, 
        cbar_kws={ 'shrink' : .9 }, 
        ax=ax, 
        annot = True, 
        annot_kws = { 'fontsize' : 12 }
    )
plot_correlation_map(preprocess_data)
#preprocess_data.corr()

The above correlation plot shows relation between several numerical features. The highest correlation is shown by pdays and previous features.

In [39]:
temp1 = preprocess_data.y.value_counts().reset_index().rename(columns={'index':'Subscription Acceptence', 'y':'Count'},)

fig = px.bar(temp1, x="Subscription Acceptence", y="Count", text = 'Count', height = 500, width = 700, 
             color_discrete_sequence=['rgb(0,134,139)', 'rgb(231,63,116)'])
# fig.update_traces(marker_color='rgb(158,202,225)', marker_line_color='rgb(8,48,107)',
#                   marker_line_width=1.5, opacity=0.6)
fig.update_traces(texttemplate='%{text:.2s}', textposition='outside')
fig.show()

The number of observations of calls with the result NO is around 8 times higher than YES. This difference between the number of observations in the target feature is known as Imbalance Dataset.

Predictive Analysis¶

For predictive analysis, we are going to use 2 machine learning models, Catboost and XGBoost.

->The reason behind using catboost is that since we have quite a number of categorical features, we have to use encoding to convert them to numerical features which is automatically handled by catboost.

-> When we have a lot o numerical columns, it is advised to use boosting algorithms like LGB or XGB and for bagging we can use Random Forest. After trying all three of the algorithms mentioned above, we came to a conclusion of using XGBoost because it's more accurate on unforseen data.

After getting results separately from Catboost and XGBoost, we have ensembled the results and have taken the average of the results to reach the final conclusion. We are using mean_absolute_error as the metric to calculate the accuracy.

Catboost Modeling¶

In [40]:
preprocess_data.job = preprocess_data.job.str.replace('-','_')
In [41]:
## change Categorical Target column to Numerical.
preprocess_data.y = preprocess_data.y.str.replace('yes','1')
preprocess_data.y = preprocess_data.y.str.replace('no','0')

preprocess_data.y = preprocess_data.y.astype(int)
In [42]:
## Prepare a test data, which will work as unforseen data
test_data = preprocess_data.sample(4500).reset_index(drop=True)
train_data = preprocess_data.drop(test_data.index, axis=0).reset_index(drop=True)
In [43]:
## Segregate dependent variable from indepedent variables 
train_data_X = train_data.drop(['y'],axis=1)
train_data_Y = train_data.y

test_data_X = test_data.drop(['y'],axis=1)
test_data_Y = test_data.y
In [44]:
cat_features = train_data_X.select_dtypes(include=['object']).columns.tolist()

X_train, X_test, y_train, y_test = train_test_split(test_data_X,test_data_Y,train_size=.85,random_state=1234)

model = CatBoostClassifier(eval_metric='Accuracy',use_best_model=True,random_seed=42, scale_pos_weight=5)
model.fit(X_train,y_train,cat_features=cat_features,eval_set=(X_test,y_test))

y_pred = model.predict(X_test)
print(1-mean_absolute_error(y_test, y_pred))


final_pred = model.predict_proba(test_data_X)

Learning rate set to 0.044139
0:	learn: 0.7614213	test: 0.7626614	best: 0.7626614 (0)	total: 51.4ms	remaining: 51.3s
1:	learn: 0.7780501	test: 0.7507448	best: 0.7626614 (0)	total: 54ms	remaining: 27s
2:	learn: 0.7822510	test: 0.7696127	best: 0.7696127 (2)	total: 56.6ms	remaining: 18.8s
3:	learn: 0.7848766	test: 0.7656405	best: 0.7696127 (2)	total: 58.9ms	remaining: 14.7s
4:	learn: 0.7899527	test: 0.7656405	best: 0.7696127 (2)	total: 61.2ms	remaining: 12.2s
5:	learn: 0.7901278	test: 0.7656405	best: 0.7696127 (2)	total: 63.4ms	remaining: 10.5s
6:	learn: 0.7913531	test: 0.7656405	best: 0.7696127 (2)	total: 65.8ms	remaining: 9.33s
7:	learn: 0.7910030	test: 0.7626614	best: 0.7696127 (2)	total: 68ms	remaining: 8.43s
8:	learn: 0.7964292	test: 0.7646475	best: 0.7696127 (2)	total: 70.1ms	remaining: 7.71s
9:	learn: 0.8032557	test: 0.7725919	best: 0.7725919 (9)	total: 72.3ms	remaining: 7.16s
10:	learn: 0.8037809	test: 0.7686197	best: 0.7725919 (9)	total: 74.6ms	remaining: 6.7s
11:	learn: 0.8125328	test: 0.7845084	best: 0.7845084 (11)	total: 76.9ms	remaining: 6.33s
12:	learn: 0.8230352	test: 0.7894737	best: 0.7894737 (12)	total: 79.2ms	remaining: 6.01s
13:	learn: 0.8223350	test: 0.7904667	best: 0.7904667 (13)	total: 81.6ms	remaining: 5.75s
14:	learn: 0.8209347	test: 0.7904667	best: 0.7904667 (13)	total: 83.8ms	remaining: 5.5s
15:	learn: 0.8253107	test: 0.7904667	best: 0.7904667 (13)	total: 86.1ms	remaining: 5.29s
16:	learn: 0.8289865	test: 0.7914598	best: 0.7914598 (16)	total: 87.8ms	remaining: 5.08s
17:	learn: 0.8305619	test: 0.8023833	best: 0.8023833 (17)	total: 89.8ms	remaining: 4.9s
18:	learn: 0.8351129	test: 0.8023833	best: 0.8023833 (17)	total: 92.3ms	remaining: 4.77s
19:	learn: 0.8342377	test: 0.7924528	best: 0.8023833 (17)	total: 94.5ms	remaining: 4.63s
20:	learn: 0.8342377	test: 0.7914598	best: 0.8023833 (17)	total: 96.9ms	remaining: 4.51s
21:	learn: 0.8340627	test: 0.7944389	best: 0.8023833 (17)	total: 99.2ms	remaining: 4.41s
22:	learn: 0.8335375	test: 0.7934459	best: 0.8023833 (17)	total: 101ms	remaining: 4.28s
23:	learn: 0.8333625	test: 0.7974181	best: 0.8023833 (17)	total: 103ms	remaining: 4.19s
24:	learn: 0.8309120	test: 0.7964250	best: 0.8023833 (17)	total: 106ms	remaining: 4.12s
25:	learn: 0.8342377	test: 0.7954320	best: 0.8023833 (17)	total: 108ms	remaining: 4.04s
26:	learn: 0.8352879	test: 0.7914598	best: 0.8023833 (17)	total: 109ms	remaining: 3.93s
27:	learn: 0.8356380	test: 0.7994042	best: 0.8023833 (17)	total: 112ms	remaining: 3.87s
28:	learn: 0.8386137	test: 0.7914598	best: 0.8023833 (17)	total: 114ms	remaining: 3.82s
29:	learn: 0.8417644	test: 0.7974181	best: 0.8023833 (17)	total: 116ms	remaining: 3.76s
30:	learn: 0.8443900	test: 0.8053625	best: 0.8053625 (30)	total: 119ms	remaining: 3.71s
31:	learn: 0.8443900	test: 0.8123138	best: 0.8123138 (31)	total: 121ms	remaining: 3.66s
32:	learn: 0.8468405	test: 0.7994042	best: 0.8123138 (31)	total: 123ms	remaining: 3.62s
33:	learn: 0.8485909	test: 0.8083416	best: 0.8123138 (31)	total: 126ms	remaining: 3.57s
34:	learn: 0.8485909	test: 0.8043694	best: 0.8123138 (31)	total: 128ms	remaining: 3.53s
35:	learn: 0.8519167	test: 0.7954320	best: 0.8123138 (31)	total: 130ms	remaining: 3.48s
36:	learn: 0.8484159	test: 0.7984111	best: 0.8123138 (31)	total: 132ms	remaining: 3.42s
37:	learn: 0.8489410	test: 0.7954320	best: 0.8123138 (31)	total: 134ms	remaining: 3.4s
38:	learn: 0.8487660	test: 0.7914598	best: 0.8123138 (31)	total: 137ms	remaining: 3.37s
39:	learn: 0.8480658	test: 0.7944389	best: 0.8123138 (31)	total: 139ms	remaining: 3.34s
40:	learn: 0.8512165	test: 0.7984111	best: 0.8123138 (31)	total: 141ms	remaining: 3.31s
41:	learn: 0.8506914	test: 0.7924528	best: 0.8123138 (31)	total: 144ms	remaining: 3.28s
42:	learn: 0.8487660	test: 0.7924528	best: 0.8123138 (31)	total: 146ms	remaining: 3.25s
43:	learn: 0.8531420	test: 0.8013903	best: 0.8123138 (31)	total: 148ms	remaining: 3.23s
44:	learn: 0.8527919	test: 0.7994042	best: 0.8123138 (31)	total: 151ms	remaining: 3.2s
45:	learn: 0.8547173	test: 0.8033764	best: 0.8123138 (31)	total: 153ms	remaining: 3.18s
46:	learn: 0.8568178	test: 0.8033764	best: 0.8123138 (31)	total: 156ms	remaining: 3.15s
47:	learn: 0.8562927	test: 0.8053625	best: 0.8123138 (31)	total: 158ms	remaining: 3.13s
48:	learn: 0.8571679	test: 0.8003972	best: 0.8123138 (31)	total: 160ms	remaining: 3.11s
49:	learn: 0.8585682	test: 0.8123138	best: 0.8123138 (31)	total: 163ms	remaining: 3.09s
50:	learn: 0.8587432	test: 0.8142999	best: 0.8142999 (50)	total: 165ms	remaining: 3.07s
51:	learn: 0.8611938	test: 0.8083416	best: 0.8142999 (50)	total: 167ms	remaining: 3.05s
52:	learn: 0.8592683	test: 0.8063555	best: 0.8142999 (50)	total: 169ms	remaining: 3.02s
53:	learn: 0.8592683	test: 0.8013903	best: 0.8142999 (50)	total: 172ms	remaining: 3.01s
54:	learn: 0.8618939	test: 0.7954320	best: 0.8142999 (50)	total: 174ms	remaining: 2.99s
55:	learn: 0.8625941	test: 0.7964250	best: 0.8142999 (50)	total: 177ms	remaining: 2.98s
56:	learn: 0.8634693	test: 0.8013903	best: 0.8142999 (50)	total: 179ms	remaining: 2.96s
57:	learn: 0.8646946	test: 0.8023833	best: 0.8142999 (50)	total: 180ms	remaining: 2.93s
58:	learn: 0.8652197	test: 0.8103277	best: 0.8142999 (50)	total: 183ms	remaining: 2.92s
59:	learn: 0.8643445	test: 0.8113208	best: 0.8142999 (50)	total: 185ms	remaining: 2.9s
60:	learn: 0.8657448	test: 0.8113208	best: 0.8142999 (50)	total: 187ms	remaining: 2.88s
61:	learn: 0.8659198	test: 0.8053625	best: 0.8142999 (50)	total: 190ms	remaining: 2.87s
62:	learn: 0.8659198	test: 0.8053625	best: 0.8142999 (50)	total: 192ms	remaining: 2.86s
63:	learn: 0.8650446	test: 0.8073486	best: 0.8142999 (50)	total: 195ms	remaining: 2.85s
64:	learn: 0.8655698	test: 0.8123138	best: 0.8142999 (50)	total: 198ms	remaining: 2.84s
65:	learn: 0.8662699	test: 0.8073486	best: 0.8142999 (50)	total: 200ms	remaining: 2.83s
66:	learn: 0.8659198	test: 0.8023833	best: 0.8142999 (50)	total: 204ms	remaining: 2.84s
67:	learn: 0.8702958	test: 0.8033764	best: 0.8142999 (50)	total: 207ms	remaining: 2.83s
68:	learn: 0.8718712	test: 0.8083416	best: 0.8142999 (50)	total: 210ms	remaining: 2.83s
69:	learn: 0.8704709	test: 0.8033764	best: 0.8142999 (50)	total: 212ms	remaining: 2.82s
70:	learn: 0.8732715	test: 0.8083416	best: 0.8142999 (50)	total: 215ms	remaining: 2.81s
71:	learn: 0.8730964	test: 0.8083416	best: 0.8142999 (50)	total: 216ms	remaining: 2.79s
72:	learn: 0.8715211	test: 0.8033764	best: 0.8142999 (50)	total: 219ms	remaining: 2.78s
73:	learn: 0.8709960	test: 0.8232373	best: 0.8232373 (73)	total: 221ms	remaining: 2.77s
74:	learn: 0.8739716	test: 0.8301887	best: 0.8301887 (74)	total: 224ms	remaining: 2.76s
75:	learn: 0.8748468	test: 0.8202582	best: 0.8301887 (74)	total: 226ms	remaining: 2.75s
76:	learn: 0.8737966	test: 0.8212512	best: 0.8301887 (74)	total: 229ms	remaining: 2.74s
77:	learn: 0.8727464	test: 0.8202582	best: 0.8301887 (74)	total: 231ms	remaining: 2.73s
78:	learn: 0.8739716	test: 0.8142999	best: 0.8301887 (74)	total: 234ms	remaining: 2.72s
79:	learn: 0.8741467	test: 0.8192651	best: 0.8301887 (74)	total: 236ms	remaining: 2.71s
80:	learn: 0.8764222	test: 0.8252234	best: 0.8301887 (74)	total: 238ms	remaining: 2.7s
81:	learn: 0.8755470	test: 0.8192651	best: 0.8301887 (74)	total: 240ms	remaining: 2.69s
82:	learn: 0.8760721	test: 0.8242304	best: 0.8301887 (74)	total: 243ms	remaining: 2.68s
83:	learn: 0.8746718	test: 0.8192651	best: 0.8301887 (74)	total: 245ms	remaining: 2.67s
84:	learn: 0.8748468	test: 0.8212512	best: 0.8301887 (74)	total: 247ms	remaining: 2.66s
85:	learn: 0.8769473	test: 0.8272095	best: 0.8301887 (74)	total: 250ms	remaining: 2.66s
86:	learn: 0.8765972	test: 0.8282026	best: 0.8301887 (74)	total: 253ms	remaining: 2.65s
87:	learn: 0.8764222	test: 0.8282026	best: 0.8301887 (74)	total: 255ms	remaining: 2.64s
88:	learn: 0.8781726	test: 0.8272095	best: 0.8301887 (74)	total: 258ms	remaining: 2.64s
89:	learn: 0.8799230	test: 0.8282026	best: 0.8301887 (74)	total: 260ms	remaining: 2.63s
90:	learn: 0.8783476	test: 0.8291956	best: 0.8301887 (74)	total: 262ms	remaining: 2.62s
91:	learn: 0.8786977	test: 0.8291956	best: 0.8301887 (74)	total: 265ms	remaining: 2.61s
92:	learn: 0.8793979	test: 0.8222443	best: 0.8301887 (74)	total: 267ms	remaining: 2.6s
93:	learn: 0.8799230	test: 0.8222443	best: 0.8301887 (74)	total: 269ms	remaining: 2.6s
94:	learn: 0.8807982	test: 0.8222443	best: 0.8301887 (74)	total: 271ms	remaining: 2.58s
95:	learn: 0.8818484	test: 0.8232373	best: 0.8301887 (74)	total: 273ms	remaining: 2.57s
96:	learn: 0.8809732	test: 0.8232373	best: 0.8301887 (74)	total: 275ms	remaining: 2.56s
97:	learn: 0.8816734	test: 0.8242304	best: 0.8301887 (74)	total: 278ms	remaining: 2.56s
98:	learn: 0.8814983	test: 0.8232373	best: 0.8301887 (74)	total: 279ms	remaining: 2.54s
99:	learn: 0.8823735	test: 0.8222443	best: 0.8301887 (74)	total: 281ms	remaining: 2.53s
100:	learn: 0.8825486	test: 0.8222443	best: 0.8301887 (74)	total: 284ms	remaining: 2.52s
101:	learn: 0.8827236	test: 0.8222443	best: 0.8301887 (74)	total: 286ms	remaining: 2.52s
102:	learn: 0.8820235	test: 0.8212512	best: 0.8301887 (74)	total: 289ms	remaining: 2.51s
103:	learn: 0.8818484	test: 0.8232373	best: 0.8301887 (74)	total: 291ms	remaining: 2.51s
104:	learn: 0.8818484	test: 0.8222443	best: 0.8301887 (74)	total: 294ms	remaining: 2.5s
105:	learn: 0.8816734	test: 0.8232373	best: 0.8301887 (74)	total: 296ms	remaining: 2.5s
106:	learn: 0.8816734	test: 0.8232373	best: 0.8301887 (74)	total: 299ms	remaining: 2.49s
107:	learn: 0.8816734	test: 0.8232373	best: 0.8301887 (74)	total: 301ms	remaining: 2.49s
108:	learn: 0.8835988	test: 0.8212512	best: 0.8301887 (74)	total: 304ms	remaining: 2.48s
109:	learn: 0.8842990	test: 0.8212512	best: 0.8301887 (74)	total: 307ms	remaining: 2.48s
110:	learn: 0.8842990	test: 0.8262165	best: 0.8301887 (74)	total: 309ms	remaining: 2.48s
111:	learn: 0.8842990	test: 0.8262165	best: 0.8301887 (74)	total: 311ms	remaining: 2.47s
112:	learn: 0.8858743	test: 0.8262165	best: 0.8301887 (74)	total: 314ms	remaining: 2.46s
113:	learn: 0.8848241	test: 0.8272095	best: 0.8301887 (74)	total: 315ms	remaining: 2.45s
114:	learn: 0.8839489	test: 0.8262165	best: 0.8301887 (74)	total: 318ms	remaining: 2.44s
115:	learn: 0.8858743	test: 0.8262165	best: 0.8301887 (74)	total: 320ms	remaining: 2.44s
116:	learn: 0.8853492	test: 0.8272095	best: 0.8301887 (74)	total: 323ms	remaining: 2.44s
117:	learn: 0.8855242	test: 0.8222443	best: 0.8301887 (74)	total: 325ms	remaining: 2.43s
118:	learn: 0.8855242	test: 0.8222443	best: 0.8301887 (74)	total: 327ms	remaining: 2.42s
119:	learn: 0.8877998	test: 0.8272095	best: 0.8301887 (74)	total: 329ms	remaining: 2.42s
120:	learn: 0.8890250	test: 0.8272095	best: 0.8301887 (74)	total: 332ms	remaining: 2.41s
121:	learn: 0.8869246	test: 0.8272095	best: 0.8301887 (74)	total: 334ms	remaining: 2.41s
122:	learn: 0.8884999	test: 0.8222443	best: 0.8301887 (74)	total: 337ms	remaining: 2.4s
123:	learn: 0.8881498	test: 0.8222443	best: 0.8301887 (74)	total: 339ms	remaining: 2.4s
124:	learn: 0.8884999	test: 0.8222443	best: 0.8301887 (74)	total: 342ms	remaining: 2.39s
125:	learn: 0.8884999	test: 0.8222443	best: 0.8301887 (74)	total: 345ms	remaining: 2.39s
126:	learn: 0.8884999	test: 0.8222443	best: 0.8301887 (74)	total: 346ms	remaining: 2.38s
127:	learn: 0.8877998	test: 0.8222443	best: 0.8301887 (74)	total: 348ms	remaining: 2.37s
128:	learn: 0.8881498	test: 0.8212512	best: 0.8301887 (74)	total: 351ms	remaining: 2.37s
129:	learn: 0.8890250	test: 0.8212512	best: 0.8301887 (74)	total: 353ms	remaining: 2.36s
130:	learn: 0.8892001	test: 0.8212512	best: 0.8301887 (74)	total: 356ms	remaining: 2.36s
131:	learn: 0.8890250	test: 0.8212512	best: 0.8301887 (74)	total: 357ms	remaining: 2.35s
132:	learn: 0.8890250	test: 0.8212512	best: 0.8301887 (74)	total: 359ms	remaining: 2.34s
133:	learn: 0.8893751	test: 0.8212512	best: 0.8301887 (74)	total: 362ms	remaining: 2.34s
134:	learn: 0.8920007	test: 0.8222443	best: 0.8301887 (74)	total: 365ms	remaining: 2.34s
135:	learn: 0.8909505	test: 0.8232373	best: 0.8301887 (74)	total: 366ms	remaining: 2.33s
136:	learn: 0.8914756	test: 0.8222443	best: 0.8301887 (74)	total: 369ms	remaining: 2.32s
137:	learn: 0.8900753	test: 0.8232373	best: 0.8301887 (74)	total: 371ms	remaining: 2.32s
138:	learn: 0.8913005	test: 0.8242304	best: 0.8301887 (74)	total: 374ms	remaining: 2.32s
139:	learn: 0.8927009	test: 0.8232373	best: 0.8301887 (74)	total: 376ms	remaining: 2.31s
140:	learn: 0.8935761	test: 0.8162860	best: 0.8301887 (74)	total: 379ms	remaining: 2.31s
141:	learn: 0.8935761	test: 0.8162860	best: 0.8301887 (74)	total: 380ms	remaining: 2.3s
142:	learn: 0.8932260	test: 0.8172790	best: 0.8301887 (74)	total: 383ms	remaining: 2.29s
143:	learn: 0.8928759	test: 0.8152929	best: 0.8301887 (74)	total: 385ms	remaining: 2.29s
144:	learn: 0.8930509	test: 0.8162860	best: 0.8301887 (74)	total: 387ms	remaining: 2.28s
145:	learn: 0.8944513	test: 0.8172790	best: 0.8301887 (74)	total: 390ms	remaining: 2.28s
146:	learn: 0.8941012	test: 0.8172790	best: 0.8301887 (74)	total: 393ms	remaining: 2.28s
147:	learn: 0.8944513	test: 0.8172790	best: 0.8301887 (74)	total: 396ms	remaining: 2.28s
148:	learn: 0.8944513	test: 0.8172790	best: 0.8301887 (74)	total: 397ms	remaining: 2.27s
149:	learn: 0.8948013	test: 0.8152929	best: 0.8301887 (74)	total: 400ms	remaining: 2.27s
150:	learn: 0.8942762	test: 0.8172790	best: 0.8301887 (74)	total: 403ms	remaining: 2.26s
151:	learn: 0.8958516	test: 0.8182721	best: 0.8301887 (74)	total: 406ms	remaining: 2.27s
152:	learn: 0.8944513	test: 0.8182721	best: 0.8301887 (74)	total: 410ms	remaining: 2.27s
153:	learn: 0.8946263	test: 0.8182721	best: 0.8301887 (74)	total: 411ms	remaining: 2.26s
154:	learn: 0.8956765	test: 0.8182721	best: 0.8301887 (74)	total: 414ms	remaining: 2.26s
155:	learn: 0.8953264	test: 0.8182721	best: 0.8301887 (74)	total: 417ms	remaining: 2.26s
156:	learn: 0.8944513	test: 0.8182721	best: 0.8301887 (74)	total: 419ms	remaining: 2.25s
157:	learn: 0.8949764	test: 0.8172790	best: 0.8301887 (74)	total: 422ms	remaining: 2.25s
158:	learn: 0.8948013	test: 0.8162860	best: 0.8301887 (74)	total: 425ms	remaining: 2.25s
159:	learn: 0.8948013	test: 0.8172790	best: 0.8301887 (74)	total: 427ms	remaining: 2.24s
160:	learn: 0.8949764	test: 0.8182721	best: 0.8301887 (74)	total: 430ms	remaining: 2.24s
161:	learn: 0.8941012	test: 0.8182721	best: 0.8301887 (74)	total: 432ms	remaining: 2.23s
162:	learn: 0.8941012	test: 0.8192651	best: 0.8301887 (74)	total: 435ms	remaining: 2.23s
163:	learn: 0.8951514	test: 0.8172790	best: 0.8301887 (74)	total: 439ms	remaining: 2.23s
164:	learn: 0.8951514	test: 0.8172790	best: 0.8301887 (74)	total: 440ms	remaining: 2.22s
165:	learn: 0.8948013	test: 0.8192651	best: 0.8301887 (74)	total: 442ms	remaining: 2.22s
166:	learn: 0.8963767	test: 0.8222443	best: 0.8301887 (74)	total: 445ms	remaining: 2.22s
167:	learn: 0.8976020	test: 0.8172790	best: 0.8301887 (74)	total: 447ms	remaining: 2.21s
168:	learn: 0.8986522	test: 0.8172790	best: 0.8301887 (74)	total: 450ms	remaining: 2.21s
169:	learn: 0.8986522	test: 0.8172790	best: 0.8301887 (74)	total: 453ms	remaining: 2.21s
170:	learn: 0.8967268	test: 0.8172790	best: 0.8301887 (74)	total: 455ms	remaining: 2.21s
171:	learn: 0.8967268	test: 0.8172790	best: 0.8301887 (74)	total: 456ms	remaining: 2.2s
172:	learn: 0.8986522	test: 0.8172790	best: 0.8301887 (74)	total: 459ms	remaining: 2.19s
173:	learn: 0.8976020	test: 0.8162860	best: 0.8301887 (74)	total: 461ms	remaining: 2.19s
174:	learn: 0.8972519	test: 0.8162860	best: 0.8301887 (74)	total: 463ms	remaining: 2.18s
175:	learn: 0.8998775	test: 0.8212512	best: 0.8301887 (74)	total: 466ms	remaining: 2.18s
176:	learn: 0.8997024	test: 0.8232373	best: 0.8301887 (74)	total: 468ms	remaining: 2.18s
177:	learn: 0.9004026	test: 0.8232373	best: 0.8301887 (74)	total: 473ms	remaining: 2.18s
178:	learn: 0.9002276	test: 0.8242304	best: 0.8301887 (74)	total: 475ms	remaining: 2.18s
179:	learn: 0.9004026	test: 0.8242304	best: 0.8301887 (74)	total: 477ms	remaining: 2.17s
180:	learn: 0.9004026	test: 0.8242304	best: 0.8301887 (74)	total: 479ms	remaining: 2.17s
181:	learn: 0.9004026	test: 0.8242304	best: 0.8301887 (74)	total: 480ms	remaining: 2.16s
182:	learn: 0.9026781	test: 0.8232373	best: 0.8301887 (74)	total: 483ms	remaining: 2.16s
183:	learn: 0.9035533	test: 0.8242304	best: 0.8301887 (74)	total: 486ms	remaining: 2.15s
184:	learn: 0.9014528	test: 0.8202582	best: 0.8301887 (74)	total: 488ms	remaining: 2.15s
185:	learn: 0.9018029	test: 0.8252234	best: 0.8301887 (74)	total: 491ms	remaining: 2.15s
186:	learn: 0.9028531	test: 0.8252234	best: 0.8301887 (74)	total: 493ms	remaining: 2.15s
187:	learn: 0.9023280	test: 0.8242304	best: 0.8301887 (74)	total: 496ms	remaining: 2.14s
188:	learn: 0.9023280	test: 0.8242304	best: 0.8301887 (74)	total: 497ms	remaining: 2.13s
189:	learn: 0.9035533	test: 0.8202582	best: 0.8301887 (74)	total: 500ms	remaining: 2.13s
190:	learn: 0.9037283	test: 0.8202582	best: 0.8301887 (74)	total: 502ms	remaining: 2.13s
191:	learn: 0.9037283	test: 0.8202582	best: 0.8301887 (74)	total: 504ms	remaining: 2.12s
192:	learn: 0.9047786	test: 0.8212512	best: 0.8301887 (74)	total: 507ms	remaining: 2.12s
193:	learn: 0.9037283	test: 0.8222443	best: 0.8301887 (74)	total: 509ms	remaining: 2.12s
194:	learn: 0.9037283	test: 0.8202582	best: 0.8301887 (74)	total: 512ms	remaining: 2.11s
195:	learn: 0.9039034	test: 0.8212512	best: 0.8301887 (74)	total: 515ms	remaining: 2.11s
196:	learn: 0.9086294	test: 0.8192651	best: 0.8301887 (74)	total: 517ms	remaining: 2.11s
197:	learn: 0.9063539	test: 0.8202582	best: 0.8301887 (74)	total: 520ms	remaining: 2.1s
198:	learn: 0.9072291	test: 0.8202582	best: 0.8301887 (74)	total: 523ms	remaining: 2.1s
199:	learn: 0.9061789	test: 0.8202582	best: 0.8301887 (74)	total: 525ms	remaining: 2.1s
200:	learn: 0.9093296	test: 0.8202582	best: 0.8301887 (74)	total: 528ms	remaining: 2.1s
201:	learn: 0.9095046	test: 0.8202582	best: 0.8301887 (74)	total: 531ms	remaining: 2.1s
202:	learn: 0.9095046	test: 0.8202582	best: 0.8301887 (74)	total: 533ms	remaining: 2.09s
203:	learn: 0.9095046	test: 0.8162860	best: 0.8301887 (74)	total: 536ms	remaining: 2.09s
204:	learn: 0.9100298	test: 0.8133069	best: 0.8301887 (74)	total: 538ms	remaining: 2.09s
205:	learn: 0.9114301	test: 0.8083416	best: 0.8301887 (74)	total: 541ms	remaining: 2.08s
206:	learn: 0.9114301	test: 0.8093347	best: 0.8301887 (74)	total: 544ms	remaining: 2.08s
207:	learn: 0.9121302	test: 0.8093347	best: 0.8301887 (74)	total: 546ms	remaining: 2.08s
208:	learn: 0.9119552	test: 0.8083416	best: 0.8301887 (74)	total: 549ms	remaining: 2.08s
209:	learn: 0.9109050	test: 0.8093347	best: 0.8301887 (74)	total: 552ms	remaining: 2.07s
210:	learn: 0.9107299	test: 0.8093347	best: 0.8301887 (74)	total: 554ms	remaining: 2.07s
211:	learn: 0.9109050	test: 0.8093347	best: 0.8301887 (74)	total: 557ms	remaining: 2.07s
212:	learn: 0.9116051	test: 0.8083416	best: 0.8301887 (74)	total: 559ms	remaining: 2.07s
213:	learn: 0.9126553	test: 0.8113208	best: 0.8301887 (74)	total: 562ms	remaining: 2.06s
214:	learn: 0.9130054	test: 0.8053625	best: 0.8301887 (74)	total: 564ms	remaining: 2.06s
215:	learn: 0.9130054	test: 0.8053625	best: 0.8301887 (74)	total: 566ms	remaining: 2.05s
216:	learn: 0.9128304	test: 0.8063555	best: 0.8301887 (74)	total: 569ms	remaining: 2.05s
217:	learn: 0.9128304	test: 0.8063555	best: 0.8301887 (74)	total: 571ms	remaining: 2.05s
218:	learn: 0.9137056	test: 0.8053625	best: 0.8301887 (74)	total: 574ms	remaining: 2.04s
219:	learn: 0.9137056	test: 0.8053625	best: 0.8301887 (74)	total: 576ms	remaining: 2.04s
220:	learn: 0.9130054	test: 0.8053625	best: 0.8301887 (74)	total: 579ms	remaining: 2.04s
221:	learn: 0.9128304	test: 0.8063555	best: 0.8301887 (74)	total: 581ms	remaining: 2.04s
222:	learn: 0.9121302	test: 0.8063555	best: 0.8301887 (74)	total: 584ms	remaining: 2.03s
223:	learn: 0.9145808	test: 0.8073486	best: 0.8301887 (74)	total: 587ms	remaining: 2.03s
224:	learn: 0.9145808	test: 0.8073486	best: 0.8301887 (74)	total: 590ms	remaining: 2.03s
225:	learn: 0.9142307	test: 0.8063555	best: 0.8301887 (74)	total: 593ms	remaining: 2.03s
226:	learn: 0.9152809	test: 0.8083416	best: 0.8301887 (74)	total: 596ms	remaining: 2.03s
227:	learn: 0.9166813	test: 0.8043694	best: 0.8301887 (74)	total: 599ms	remaining: 2.03s
228:	learn: 0.9159811	test: 0.8043694	best: 0.8301887 (74)	total: 602ms	remaining: 2.02s
229:	learn: 0.9168563	test: 0.8043694	best: 0.8301887 (74)	total: 605ms	remaining: 2.02s
230:	learn: 0.9168563	test: 0.8043694	best: 0.8301887 (74)	total: 608ms	remaining: 2.02s
231:	learn: 0.9172064	test: 0.8043694	best: 0.8301887 (74)	total: 610ms	remaining: 2.02s
232:	learn: 0.9173814	test: 0.8043694	best: 0.8301887 (74)	total: 614ms	remaining: 2.02s
233:	learn: 0.9170313	test: 0.8043694	best: 0.8301887 (74)	total: 617ms	remaining: 2.02s
234:	learn: 0.9177315	test: 0.8043694	best: 0.8301887 (74)	total: 620ms	remaining: 2.02s
235:	learn: 0.9175565	test: 0.8053625	best: 0.8301887 (74)	total: 623ms	remaining: 2.02s
236:	learn: 0.9175565	test: 0.8053625	best: 0.8301887 (74)	total: 624ms	remaining: 2.01s
237:	learn: 0.9173814	test: 0.8063555	best: 0.8301887 (74)	total: 627ms	remaining: 2.01s
238:	learn: 0.9170313	test: 0.8053625	best: 0.8301887 (74)	total: 630ms	remaining: 2s
239:	learn: 0.9173814	test: 0.7994042	best: 0.8301887 (74)	total: 632ms	remaining: 2s
240:	learn: 0.9173814	test: 0.8043694	best: 0.8301887 (74)	total: 634ms	remaining: 2s
241:	learn: 0.9180816	test: 0.8033764	best: 0.8301887 (74)	total: 637ms	remaining: 1.99s
242:	learn: 0.9182566	test: 0.8033764	best: 0.8301887 (74)	total: 639ms	remaining: 1.99s
243:	learn: 0.9175565	test: 0.8033764	best: 0.8301887 (74)	total: 642ms	remaining: 1.99s
244:	learn: 0.9175565	test: 0.7994042	best: 0.8301887 (74)	total: 644ms	remaining: 1.99s
245:	learn: 0.9175565	test: 0.7994042	best: 0.8301887 (74)	total: 647ms	remaining: 1.98s
246:	learn: 0.9175565	test: 0.7994042	best: 0.8301887 (74)	total: 650ms	remaining: 1.98s
247:	learn: 0.9179065	test: 0.7994042	best: 0.8301887 (74)	total: 652ms	remaining: 1.98s
248:	learn: 0.9191318	test: 0.7994042	best: 0.8301887 (74)	total: 655ms	remaining: 1.97s
249:	learn: 0.9189568	test: 0.8003972	best: 0.8301887 (74)	total: 657ms	remaining: 1.97s
250:	learn: 0.9194819	test: 0.8003972	best: 0.8301887 (74)	total: 660ms	remaining: 1.97s
251:	learn: 0.9194819	test: 0.8003972	best: 0.8301887 (74)	total: 662ms	remaining: 1.96s
252:	learn: 0.9194819	test: 0.8003972	best: 0.8301887 (74)	total: 664ms	remaining: 1.96s
253:	learn: 0.9191318	test: 0.8003972	best: 0.8301887 (74)	total: 667ms	remaining: 1.96s
254:	learn: 0.9191318	test: 0.8003972	best: 0.8301887 (74)	total: 670ms	remaining: 1.96s
255:	learn: 0.9196569	test: 0.7994042	best: 0.8301887 (74)	total: 672ms	remaining: 1.95s
256:	learn: 0.9200070	test: 0.8013903	best: 0.8301887 (74)	total: 675ms	remaining: 1.95s
257:	learn: 0.9200070	test: 0.8013903	best: 0.8301887 (74)	total: 677ms	remaining: 1.95s
258:	learn: 0.9200070	test: 0.8013903	best: 0.8301887 (74)	total: 680ms	remaining: 1.95s
259:	learn: 0.9203571	test: 0.8003972	best: 0.8301887 (74)	total: 682ms	remaining: 1.94s
260:	learn: 0.9203571	test: 0.8003972	best: 0.8301887 (74)	total: 685ms	remaining: 1.94s
261:	learn: 0.9212323	test: 0.8043694	best: 0.8301887 (74)	total: 687ms	remaining: 1.94s
262:	learn: 0.9201820	test: 0.8053625	best: 0.8301887 (74)	total: 690ms	remaining: 1.93s
263:	learn: 0.9203571	test: 0.8053625	best: 0.8301887 (74)	total: 693ms	remaining: 1.93s
264:	learn: 0.9215824	test: 0.8043694	best: 0.8301887 (74)	total: 695ms	remaining: 1.93s
265:	learn: 0.9215824	test: 0.7994042	best: 0.8301887 (74)	total: 698ms	remaining: 1.93s
266:	learn: 0.9226326	test: 0.8023833	best: 0.8301887 (74)	total: 700ms	remaining: 1.92s
267:	learn: 0.9226326	test: 0.8023833	best: 0.8301887 (74)	total: 703ms	remaining: 1.92s
268:	learn: 0.9231577	test: 0.8063555	best: 0.8301887 (74)	total: 705ms	remaining: 1.92s
269:	learn: 0.9235078	test: 0.8053625	best: 0.8301887 (74)	total: 708ms	remaining: 1.91s
270:	learn: 0.9233327	test: 0.8063555	best: 0.8301887 (74)	total: 711ms	remaining: 1.91s
271:	learn: 0.9235078	test: 0.8063555	best: 0.8301887 (74)	total: 713ms	remaining: 1.91s
272:	learn: 0.9233327	test: 0.8063555	best: 0.8301887 (74)	total: 716ms	remaining: 1.91s
273:	learn: 0.9235078	test: 0.8063555	best: 0.8301887 (74)	total: 718ms	remaining: 1.9s
274:	learn: 0.9236828	test: 0.8063555	best: 0.8301887 (74)	total: 721ms	remaining: 1.9s
275:	learn: 0.9240329	test: 0.8063555	best: 0.8301887 (74)	total: 723ms	remaining: 1.9s
276:	learn: 0.9247331	test: 0.8073486	best: 0.8301887 (74)	total: 726ms	remaining: 1.9s
277:	learn: 0.9250831	test: 0.8073486	best: 0.8301887 (74)	total: 729ms	remaining: 1.89s
278:	learn: 0.9250831	test: 0.8023833	best: 0.8301887 (74)	total: 731ms	remaining: 1.89s
279:	learn: 0.9250831	test: 0.8023833	best: 0.8301887 (74)	total: 734ms	remaining: 1.89s
280:	learn: 0.9249081	test: 0.8023833	best: 0.8301887 (74)	total: 736ms	remaining: 1.88s
281:	learn: 0.9249081	test: 0.8023833	best: 0.8301887 (74)	total: 739ms	remaining: 1.88s
282:	learn: 0.9256083	test: 0.7974181	best: 0.8301887 (74)	total: 741ms	remaining: 1.88s
283:	learn: 0.9252582	test: 0.7984111	best: 0.8301887 (74)	total: 744ms	remaining: 1.88s
284:	learn: 0.9256083	test: 0.8033764	best: 0.8301887 (74)	total: 747ms	remaining: 1.87s
285:	learn: 0.9256083	test: 0.8033764	best: 0.8301887 (74)	total: 750ms	remaining: 1.87s
286:	learn: 0.9261334	test: 0.8033764	best: 0.8301887 (74)	total: 753ms	remaining: 1.87s
287:	learn: 0.9263084	test: 0.8043694	best: 0.8301887 (74)	total: 755ms	remaining: 1.87s
288:	learn: 0.9270086	test: 0.8043694	best: 0.8301887 (74)	total: 759ms	remaining: 1.87s
289:	learn: 0.9271836	test: 0.8043694	best: 0.8301887 (74)	total: 762ms	remaining: 1.86s
290:	learn: 0.9289340	test: 0.8043694	best: 0.8301887 (74)	total: 765ms	remaining: 1.86s
291:	learn: 0.9287590	test: 0.8043694	best: 0.8301887 (74)	total: 768ms	remaining: 1.86s
292:	learn: 0.9285839	test: 0.8053625	best: 0.8301887 (74)	total: 771ms	remaining: 1.86s
293:	learn: 0.9284089	test: 0.8053625	best: 0.8301887 (74)	total: 774ms	remaining: 1.86s
294:	learn: 0.9284089	test: 0.8053625	best: 0.8301887 (74)	total: 777ms	remaining: 1.86s
295:	learn: 0.9284089	test: 0.8053625	best: 0.8301887 (74)	total: 780ms	remaining: 1.85s
296:	learn: 0.9282339	test: 0.8053625	best: 0.8301887 (74)	total: 782ms	remaining: 1.85s
297:	learn: 0.9280588	test: 0.8053625	best: 0.8301887 (74)	total: 786ms	remaining: 1.85s
298:	learn: 0.9273587	test: 0.8003972	best: 0.8301887 (74)	total: 789ms	remaining: 1.85s
299:	learn: 0.9277087	test: 0.8003972	best: 0.8301887 (74)	total: 792ms	remaining: 1.85s
300:	learn: 0.9289340	test: 0.8003972	best: 0.8301887 (74)	total: 795ms	remaining: 1.84s
301:	learn: 0.9291090	test: 0.8003972	best: 0.8301887 (74)	total: 797ms	remaining: 1.84s
302:	learn: 0.9291090	test: 0.8003972	best: 0.8301887 (74)	total: 800ms	remaining: 1.84s
303:	learn: 0.9296342	test: 0.8003972	best: 0.8301887 (74)	total: 803ms	remaining: 1.84s
304:	learn: 0.9305094	test: 0.8003972	best: 0.8301887 (74)	total: 805ms	remaining: 1.83s
305:	learn: 0.9294591	test: 0.8013903	best: 0.8301887 (74)	total: 809ms	remaining: 1.83s
306:	learn: 0.9310345	test: 0.8003972	best: 0.8301887 (74)	total: 812ms	remaining: 1.83s
307:	learn: 0.9301593	test: 0.8003972	best: 0.8301887 (74)	total: 815ms	remaining: 1.83s
308:	learn: 0.9299842	test: 0.8043694	best: 0.8301887 (74)	total: 818ms	remaining: 1.83s
309:	learn: 0.9301593	test: 0.8043694	best: 0.8301887 (74)	total: 821ms	remaining: 1.83s
310:	learn: 0.9310345	test: 0.8003972	best: 0.8301887 (74)	total: 824ms	remaining: 1.82s
311:	learn: 0.9308594	test: 0.7944389	best: 0.8301887 (74)	total: 827ms	remaining: 1.82s
312:	learn: 0.9305094	test: 0.7944389	best: 0.8301887 (74)	total: 830ms	remaining: 1.82s
313:	learn: 0.9308594	test: 0.7944389	best: 0.8301887 (74)	total: 833ms	remaining: 1.82s
314:	learn: 0.9308594	test: 0.7954320	best: 0.8301887 (74)	total: 835ms	remaining: 1.82s
315:	learn: 0.9320847	test: 0.7964250	best: 0.8301887 (74)	total: 838ms	remaining: 1.81s
316:	learn: 0.9319097	test: 0.7964250	best: 0.8301887 (74)	total: 841ms	remaining: 1.81s
317:	learn: 0.9317346	test: 0.7964250	best: 0.8301887 (74)	total: 843ms	remaining: 1.81s
318:	learn: 0.9317346	test: 0.7964250	best: 0.8301887 (74)	total: 846ms	remaining: 1.81s
319:	learn: 0.9317346	test: 0.7964250	best: 0.8301887 (74)	total: 849ms	remaining: 1.8s
320:	learn: 0.9329599	test: 0.7974181	best: 0.8301887 (74)	total: 852ms	remaining: 1.8s
321:	learn: 0.9333100	test: 0.7964250	best: 0.8301887 (74)	total: 855ms	remaining: 1.8s
322:	learn: 0.9331350	test: 0.7974181	best: 0.8301887 (74)	total: 858ms	remaining: 1.8s
323:	learn: 0.9326098	test: 0.7964250	best: 0.8301887 (74)	total: 861ms	remaining: 1.79s
324:	learn: 0.9326098	test: 0.7964250	best: 0.8301887 (74)	total: 863ms	remaining: 1.79s
325:	learn: 0.9326098	test: 0.7964250	best: 0.8301887 (74)	total: 866ms	remaining: 1.79s
326:	learn: 0.9326098	test: 0.7964250	best: 0.8301887 (74)	total: 868ms	remaining: 1.79s
327:	learn: 0.9334850	test: 0.7964250	best: 0.8301887 (74)	total: 871ms	remaining: 1.78s
328:	learn: 0.9336601	test: 0.7964250	best: 0.8301887 (74)	total: 874ms	remaining: 1.78s
329:	learn: 0.9333100	test: 0.7964250	best: 0.8301887 (74)	total: 877ms	remaining: 1.78s
330:	learn: 0.9333100	test: 0.7964250	best: 0.8301887 (74)	total: 880ms	remaining: 1.78s
331:	learn: 0.9334850	test: 0.7964250	best: 0.8301887 (74)	total: 882ms	remaining: 1.77s
332:	learn: 0.9333100	test: 0.7964250	best: 0.8301887 (74)	total: 885ms	remaining: 1.77s
333:	learn: 0.9336601	test: 0.7964250	best: 0.8301887 (74)	total: 888ms	remaining: 1.77s
334:	learn: 0.9345353	test: 0.7964250	best: 0.8301887 (74)	total: 890ms	remaining: 1.77s
335:	learn: 0.9343602	test: 0.7914598	best: 0.8301887 (74)	total: 893ms	remaining: 1.76s
336:	learn: 0.9347103	test: 0.7914598	best: 0.8301887 (74)	total: 896ms	remaining: 1.76s
337:	learn: 0.9347103	test: 0.7964250	best: 0.8301887 (74)	total: 899ms	remaining: 1.76s
338:	learn: 0.9361106	test: 0.7974181	best: 0.8301887 (74)	total: 901ms	remaining: 1.76s
339:	learn: 0.9364607	test: 0.7974181	best: 0.8301887 (74)	total: 904ms	remaining: 1.75s
340:	learn: 0.9362857	test: 0.7974181	best: 0.8301887 (74)	total: 907ms	remaining: 1.75s
341:	learn: 0.9362857	test: 0.7974181	best: 0.8301887 (74)	total: 909ms	remaining: 1.75s
342:	learn: 0.9373359	test: 0.7974181	best: 0.8301887 (74)	total: 912ms	remaining: 1.75s
343:	learn: 0.9366357	test: 0.7974181	best: 0.8301887 (74)	total: 915ms	remaining: 1.74s
344:	learn: 0.9364607	test: 0.7974181	best: 0.8301887 (74)	total: 917ms	remaining: 1.74s
345:	learn: 0.9373359	test: 0.7974181	best: 0.8301887 (74)	total: 920ms	remaining: 1.74s
346:	learn: 0.9368108	test: 0.7974181	best: 0.8301887 (74)	total: 923ms	remaining: 1.74s
347:	learn: 0.9368108	test: 0.7974181	best: 0.8301887 (74)	total: 926ms	remaining: 1.73s
348:	learn: 0.9366357	test: 0.7984111	best: 0.8301887 (74)	total: 928ms	remaining: 1.73s
349:	learn: 0.9366357	test: 0.7984111	best: 0.8301887 (74)	total: 931ms	remaining: 1.73s
350:	learn: 0.9371609	test: 0.7974181	best: 0.8301887 (74)	total: 933ms	remaining: 1.73s
351:	learn: 0.9369858	test: 0.7974181	best: 0.8301887 (74)	total: 936ms	remaining: 1.72s
352:	learn: 0.9382111	test: 0.7974181	best: 0.8301887 (74)	total: 939ms	remaining: 1.72s
353:	learn: 0.9382111	test: 0.7984111	best: 0.8301887 (74)	total: 942ms	remaining: 1.72s
354:	learn: 0.9382111	test: 0.8033764	best: 0.8301887 (74)	total: 944ms	remaining: 1.72s
355:	learn: 0.9387362	test: 0.8033764	best: 0.8301887 (74)	total: 947ms	remaining: 1.71s
356:	learn: 0.9376860	test: 0.8043694	best: 0.8301887 (74)	total: 949ms	remaining: 1.71s
357:	learn: 0.9382111	test: 0.8043694	best: 0.8301887 (74)	total: 952ms	remaining: 1.71s
358:	learn: 0.9382111	test: 0.8043694	best: 0.8301887 (74)	total: 955ms	remaining: 1.7s
359:	learn: 0.9385612	test: 0.8043694	best: 0.8301887 (74)	total: 957ms	remaining: 1.7s
360:	learn: 0.9389113	test: 0.8053625	best: 0.8301887 (74)	total: 960ms	remaining: 1.7s
361:	learn: 0.9392613	test: 0.8043694	best: 0.8301887 (74)	total: 963ms	remaining: 1.7s
362:	learn: 0.9392613	test: 0.8043694	best: 0.8301887 (74)	total: 966ms	remaining: 1.69s
363:	learn: 0.9392613	test: 0.7984111	best: 0.8301887 (74)	total: 968ms	remaining: 1.69s
364:	learn: 0.9396114	test: 0.7934459	best: 0.8301887 (74)	total: 971ms	remaining: 1.69s
365:	learn: 0.9396114	test: 0.7984111	best: 0.8301887 (74)	total: 974ms	remaining: 1.69s
366:	learn: 0.9396114	test: 0.7984111	best: 0.8301887 (74)	total: 977ms	remaining: 1.68s
367:	learn: 0.9394364	test: 0.7994042	best: 0.8301887 (74)	total: 979ms	remaining: 1.68s
368:	learn: 0.9394364	test: 0.7944389	best: 0.8301887 (74)	total: 982ms	remaining: 1.68s
369:	learn: 0.9394364	test: 0.7954320	best: 0.8301887 (74)	total: 985ms	remaining: 1.68s
370:	learn: 0.9394364	test: 0.7954320	best: 0.8301887 (74)	total: 988ms	remaining: 1.67s
371:	learn: 0.9396114	test: 0.7954320	best: 0.8301887 (74)	total: 990ms	remaining: 1.67s
372:	learn: 0.9396114	test: 0.7964250	best: 0.8301887 (74)	total: 994ms	remaining: 1.67s
373:	learn: 0.9406616	test: 0.7954320	best: 0.8301887 (74)	total: 996ms	remaining: 1.67s
374:	learn: 0.9406616	test: 0.7964250	best: 0.8301887 (74)	total: 999ms	remaining: 1.66s
375:	learn: 0.9406616	test: 0.7964250	best: 0.8301887 (74)	total: 1s	remaining: 1.66s
376:	learn: 0.9406616	test: 0.7964250	best: 0.8301887 (74)	total: 1s	remaining: 1.66s
377:	learn: 0.9408367	test: 0.7964250	best: 0.8301887 (74)	total: 1.01s	remaining: 1.66s
378:	learn: 0.9408367	test: 0.7964250	best: 0.8301887 (74)	total: 1.01s	remaining: 1.65s
379:	learn: 0.9410117	test: 0.7964250	best: 0.8301887 (74)	total: 1.01s	remaining: 1.65s
380:	learn: 0.9410117	test: 0.7964250	best: 0.8301887 (74)	total: 1.01s	remaining: 1.65s
381:	learn: 0.9420620	test: 0.7974181	best: 0.8301887 (74)	total: 1.02s	remaining: 1.65s
382:	learn: 0.9427621	test: 0.7914598	best: 0.8301887 (74)	total: 1.02s	remaining: 1.64s
383:	learn: 0.9431122	test: 0.7914598	best: 0.8301887 (74)	total: 1.02s	remaining: 1.64s
384:	learn: 0.9418869	test: 0.7924528	best: 0.8301887 (74)	total: 1.03s	remaining: 1.64s
385:	learn: 0.9418869	test: 0.7924528	best: 0.8301887 (74)	total: 1.03s	remaining: 1.64s
386:	learn: 0.9424120	test: 0.7924528	best: 0.8301887 (74)	total: 1.03s	remaining: 1.63s
387:	learn: 0.9424120	test: 0.7924528	best: 0.8301887 (74)	total: 1.03s	remaining: 1.63s
388:	learn: 0.9424120	test: 0.7924528	best: 0.8301887 (74)	total: 1.04s	remaining: 1.63s
389:	learn: 0.9432872	test: 0.7934459	best: 0.8301887 (74)	total: 1.04s	remaining: 1.63s
390:	learn: 0.9443375	test: 0.7934459	best: 0.8301887 (74)	total: 1.04s	remaining: 1.62s
391:	learn: 0.9452127	test: 0.7934459	best: 0.8301887 (74)	total: 1.04s	remaining: 1.62s
392:	learn: 0.9452127	test: 0.7934459	best: 0.8301887 (74)	total: 1.05s	remaining: 1.62s
393:	learn: 0.9452127	test: 0.7934459	best: 0.8301887 (74)	total: 1.05s	remaining: 1.61s
394:	learn: 0.9452127	test: 0.7934459	best: 0.8301887 (74)	total: 1.05s	remaining: 1.61s
395:	learn: 0.9460879	test: 0.7924528	best: 0.8301887 (74)	total: 1.05s	remaining: 1.61s
396:	learn: 0.9452127	test: 0.7944389	best: 0.8301887 (74)	total: 1.06s	remaining: 1.61s
397:	learn: 0.9464379	test: 0.7934459	best: 0.8301887 (74)	total: 1.06s	remaining: 1.6s
398:	learn: 0.9473131	test: 0.7934459	best: 0.8301887 (74)	total: 1.06s	remaining: 1.6s
399:	learn: 0.9474882	test: 0.7934459	best: 0.8301887 (74)	total: 1.06s	remaining: 1.6s
400:	learn: 0.9476632	test: 0.7934459	best: 0.8301887 (74)	total: 1.07s	remaining: 1.59s
401:	learn: 0.9476632	test: 0.7934459	best: 0.8301887 (74)	total: 1.07s	remaining: 1.59s
402:	learn: 0.9480133	test: 0.7934459	best: 0.8301887 (74)	total: 1.07s	remaining: 1.59s
403:	learn: 0.9480133	test: 0.7934459	best: 0.8301887 (74)	total: 1.08s	remaining: 1.59s
404:	learn: 0.9481883	test: 0.7934459	best: 0.8301887 (74)	total: 1.08s	remaining: 1.58s
405:	learn: 0.9481883	test: 0.7934459	best: 0.8301887 (74)	total: 1.08s	remaining: 1.58s
406:	learn: 0.9481883	test: 0.7934459	best: 0.8301887 (74)	total: 1.08s	remaining: 1.58s
407:	learn: 0.9485384	test: 0.7934459	best: 0.8301887 (74)	total: 1.09s	remaining: 1.58s
408:	learn: 0.9485384	test: 0.7934459	best: 0.8301887 (74)	total: 1.09s	remaining: 1.57s
409:	learn: 0.9490635	test: 0.7934459	best: 0.8301887 (74)	total: 1.09s	remaining: 1.57s
410:	learn: 0.9488885	test: 0.7934459	best: 0.8301887 (74)	total: 1.09s	remaining: 1.57s
411:	learn: 0.9488885	test: 0.7934459	best: 0.8301887 (74)	total: 1.1s	remaining: 1.57s
412:	learn: 0.9501138	test: 0.7934459	best: 0.8301887 (74)	total: 1.1s	remaining: 1.56s
413:	learn: 0.9502888	test: 0.7934459	best: 0.8301887 (74)	total: 1.1s	remaining: 1.56s
414:	learn: 0.9504639	test: 0.7934459	best: 0.8301887 (74)	total: 1.11s	remaining: 1.56s
415:	learn: 0.9506389	test: 0.7934459	best: 0.8301887 (74)	total: 1.11s	remaining: 1.56s
416:	learn: 0.9494136	test: 0.7934459	best: 0.8301887 (74)	total: 1.11s	remaining: 1.55s
417:	learn: 0.9502888	test: 0.7934459	best: 0.8301887 (74)	total: 1.11s	remaining: 1.55s
418:	learn: 0.9497637	test: 0.7934459	best: 0.8301887 (74)	total: 1.12s	remaining: 1.55s
419:	learn: 0.9497637	test: 0.7924528	best: 0.8301887 (74)	total: 1.12s	remaining: 1.55s
420:	learn: 0.9499387	test: 0.7924528	best: 0.8301887 (74)	total: 1.13s	remaining: 1.55s
421:	learn: 0.9499387	test: 0.7924528	best: 0.8301887 (74)	total: 1.13s	remaining: 1.55s
422:	learn: 0.9499387	test: 0.7924528	best: 0.8301887 (74)	total: 1.13s	remaining: 1.55s
423:	learn: 0.9497637	test: 0.7924528	best: 0.8301887 (74)	total: 1.14s	remaining: 1.54s
424:	learn: 0.9497637	test: 0.7924528	best: 0.8301887 (74)	total: 1.14s	remaining: 1.54s
425:	learn: 0.9499387	test: 0.7924528	best: 0.8301887 (74)	total: 1.14s	remaining: 1.54s
426:	learn: 0.9499387	test: 0.7924528	best: 0.8301887 (74)	total: 1.15s	remaining: 1.54s
427:	learn: 0.9497637	test: 0.7874876	best: 0.8301887 (74)	total: 1.15s	remaining: 1.53s
428:	learn: 0.9497637	test: 0.7874876	best: 0.8301887 (74)	total: 1.15s	remaining: 1.53s
429:	learn: 0.9499387	test: 0.7874876	best: 0.8301887 (74)	total: 1.15s	remaining: 1.53s
430:	learn: 0.9499387	test: 0.7864945	best: 0.8301887 (74)	total: 1.16s	remaining: 1.53s
431:	learn: 0.9499387	test: 0.7874876	best: 0.8301887 (74)	total: 1.16s	remaining: 1.53s
432:	learn: 0.9497637	test: 0.7874876	best: 0.8301887 (74)	total: 1.17s	remaining: 1.52s
433:	learn: 0.9504639	test: 0.7874876	best: 0.8301887 (74)	total: 1.17s	remaining: 1.52s
434:	learn: 0.9501138	test: 0.7884806	best: 0.8301887 (74)	total: 1.17s	remaining: 1.52s
435:	learn: 0.9504639	test: 0.7884806	best: 0.8301887 (74)	total: 1.17s	remaining: 1.52s
436:	learn: 0.9509890	test: 0.7884806	best: 0.8301887 (74)	total: 1.18s	remaining: 1.52s
437:	learn: 0.9509890	test: 0.7884806	best: 0.8301887 (74)	total: 1.18s	remaining: 1.52s
438:	learn: 0.9518642	test: 0.7864945	best: 0.8301887 (74)	total: 1.18s	remaining: 1.51s
439:	learn: 0.9520392	test: 0.7864945	best: 0.8301887 (74)	total: 1.19s	remaining: 1.51s
440:	learn: 0.9532645	test: 0.7864945	best: 0.8301887 (74)	total: 1.19s	remaining: 1.51s
441:	learn: 0.9530894	test: 0.7864945	best: 0.8301887 (74)	total: 1.19s	remaining: 1.51s
442:	learn: 0.9534395	test: 0.7874876	best: 0.8301887 (74)	total: 1.2s	remaining: 1.5s
443:	learn: 0.9530894	test: 0.7864945	best: 0.8301887 (74)	total: 1.2s	remaining: 1.5s
444:	learn: 0.9534395	test: 0.7864945	best: 0.8301887 (74)	total: 1.2s	remaining: 1.5s
445:	learn: 0.9529144	test: 0.7864945	best: 0.8301887 (74)	total: 1.2s	remaining: 1.5s
446:	learn: 0.9529144	test: 0.7924528	best: 0.8301887 (74)	total: 1.21s	remaining: 1.49s
447:	learn: 0.9534395	test: 0.7914598	best: 0.8301887 (74)	total: 1.21s	remaining: 1.49s
448:	learn: 0.9532645	test: 0.7914598	best: 0.8301887 (74)	total: 1.21s	remaining: 1.49s
449:	learn: 0.9534395	test: 0.7914598	best: 0.8301887 (74)	total: 1.21s	remaining: 1.48s
450:	learn: 0.9534395	test: 0.7924528	best: 0.8301887 (74)	total: 1.22s	remaining: 1.48s
451:	learn: 0.9534395	test: 0.7924528	best: 0.8301887 (74)	total: 1.22s	remaining: 1.48s
452:	learn: 0.9543147	test: 0.7914598	best: 0.8301887 (74)	total: 1.22s	remaining: 1.48s
453:	learn: 0.9553650	test: 0.7864945	best: 0.8301887 (74)	total: 1.22s	remaining: 1.47s
454:	learn: 0.9555400	test: 0.7914598	best: 0.8301887 (74)	total: 1.23s	remaining: 1.47s
455:	learn: 0.9555400	test: 0.7914598	best: 0.8301887 (74)	total: 1.23s	remaining: 1.47s
456:	learn: 0.9555400	test: 0.7914598	best: 0.8301887 (74)	total: 1.23s	remaining: 1.46s
457:	learn: 0.9557150	test: 0.7924528	best: 0.8301887 (74)	total: 1.24s	remaining: 1.46s
458:	learn: 0.9557150	test: 0.7924528	best: 0.8301887 (74)	total: 1.24s	remaining: 1.46s
459:	learn: 0.9560651	test: 0.7924528	best: 0.8301887 (74)	total: 1.24s	remaining: 1.46s
460:	learn: 0.9560651	test: 0.7924528	best: 0.8301887 (74)	total: 1.24s	remaining: 1.45s
461:	learn: 0.9569403	test: 0.7924528	best: 0.8301887 (74)	total: 1.25s	remaining: 1.45s
462:	learn: 0.9567653	test: 0.7924528	best: 0.8301887 (74)	total: 1.25s	remaining: 1.45s
463:	learn: 0.9571154	test: 0.7914598	best: 0.8301887 (74)	total: 1.25s	remaining: 1.44s
464:	learn: 0.9569403	test: 0.7914598	best: 0.8301887 (74)	total: 1.25s	remaining: 1.44s
465:	learn: 0.9571154	test: 0.7914598	best: 0.8301887 (74)	total: 1.25s	remaining: 1.44s
466:	learn: 0.9571154	test: 0.7924528	best: 0.8301887 (74)	total: 1.26s	remaining: 1.44s
467:	learn: 0.9571154	test: 0.7924528	best: 0.8301887 (74)	total: 1.26s	remaining: 1.43s
468:	learn: 0.9581656	test: 0.7924528	best: 0.8301887 (74)	total: 1.26s	remaining: 1.43s
469:	learn: 0.9581656	test: 0.7924528	best: 0.8301887 (74)	total: 1.26s	remaining: 1.43s
470:	learn: 0.9583406	test: 0.7924528	best: 0.8301887 (74)	total: 1.27s	remaining: 1.42s
471:	learn: 0.9590408	test: 0.7924528	best: 0.8301887 (74)	total: 1.27s	remaining: 1.42s
472:	learn: 0.9590408	test: 0.7924528	best: 0.8301887 (74)	total: 1.27s	remaining: 1.42s
473:	learn: 0.9592158	test: 0.7924528	best: 0.8301887 (74)	total: 1.27s	remaining: 1.42s
474:	learn: 0.9593909	test: 0.7924528	best: 0.8301887 (74)	total: 1.28s	remaining: 1.41s
475:	learn: 0.9595659	test: 0.7924528	best: 0.8301887 (74)	total: 1.28s	remaining: 1.41s
476:	learn: 0.9597409	test: 0.7924528	best: 0.8301887 (74)	total: 1.28s	remaining: 1.41s
477:	learn: 0.9597409	test: 0.7924528	best: 0.8301887 (74)	total: 1.29s	remaining: 1.4s
478:	learn: 0.9595659	test: 0.7924528	best: 0.8301887 (74)	total: 1.29s	remaining: 1.4s
479:	learn: 0.9602661	test: 0.7924528	best: 0.8301887 (74)	total: 1.29s	remaining: 1.4s
480:	learn: 0.9604411	test: 0.7924528	best: 0.8301887 (74)	total: 1.29s	remaining: 1.4s
481:	learn: 0.9604411	test: 0.7924528	best: 0.8301887 (74)	total: 1.3s	remaining: 1.39s
482:	learn: 0.9604411	test: 0.7924528	best: 0.8301887 (74)	total: 1.3s	remaining: 1.39s
483:	learn: 0.9606161	test: 0.7924528	best: 0.8301887 (74)	total: 1.3s	remaining: 1.39s
484:	learn: 0.9606161	test: 0.7924528	best: 0.8301887 (74)	total: 1.3s	remaining: 1.38s
485:	learn: 0.9599160	test: 0.7924528	best: 0.8301887 (74)	total: 1.31s	remaining: 1.38s
486:	learn: 0.9616664	test: 0.7924528	best: 0.8301887 (74)	total: 1.31s	remaining: 1.38s
487:	learn: 0.9618414	test: 0.7924528	best: 0.8301887 (74)	total: 1.31s	remaining: 1.38s
488:	learn: 0.9620165	test: 0.7924528	best: 0.8301887 (74)	total: 1.31s	remaining: 1.37s
489:	learn: 0.9609662	test: 0.7934459	best: 0.8301887 (74)	total: 1.32s	remaining: 1.37s
490:	learn: 0.9609662	test: 0.7944389	best: 0.8301887 (74)	total: 1.32s	remaining: 1.37s
491:	learn: 0.9609662	test: 0.7944389	best: 0.8301887 (74)	total: 1.32s	remaining: 1.36s
492:	learn: 0.9609662	test: 0.7944389	best: 0.8301887 (74)	total: 1.32s	remaining: 1.36s
493:	learn: 0.9609662	test: 0.7944389	best: 0.8301887 (74)	total: 1.33s	remaining: 1.36s
494:	learn: 0.9620165	test: 0.7954320	best: 0.8301887 (74)	total: 1.33s	remaining: 1.36s
495:	learn: 0.9611413	test: 0.7954320	best: 0.8301887 (74)	total: 1.33s	remaining: 1.35s
496:	learn: 0.9620165	test: 0.7954320	best: 0.8301887 (74)	total: 1.33s	remaining: 1.35s
497:	learn: 0.9611413	test: 0.7954320	best: 0.8301887 (74)	total: 1.34s	remaining: 1.35s
498:	learn: 0.9621915	test: 0.7954320	best: 0.8301887 (74)	total: 1.34s	remaining: 1.34s
499:	learn: 0.9627166	test: 0.7954320	best: 0.8301887 (74)	total: 1.34s	remaining: 1.34s
500:	learn: 0.9627166	test: 0.7954320	best: 0.8301887 (74)	total: 1.35s	remaining: 1.34s
501:	learn: 0.9630667	test: 0.7954320	best: 0.8301887 (74)	total: 1.35s	remaining: 1.34s
502:	learn: 0.9634168	test: 0.7954320	best: 0.8301887 (74)	total: 1.35s	remaining: 1.33s
503:	learn: 0.9634168	test: 0.7954320	best: 0.8301887 (74)	total: 1.35s	remaining: 1.33s
504:	learn: 0.9634168	test: 0.7954320	best: 0.8301887 (74)	total: 1.36s	remaining: 1.33s
505:	learn: 0.9635918	test: 0.7954320	best: 0.8301887 (74)	total: 1.36s	remaining: 1.33s
506:	learn: 0.9635918	test: 0.7964250	best: 0.8301887 (74)	total: 1.36s	remaining: 1.32s
507:	learn: 0.9635918	test: 0.7964250	best: 0.8301887 (74)	total: 1.36s	remaining: 1.32s
508:	learn: 0.9635918	test: 0.7974181	best: 0.8301887 (74)	total: 1.37s	remaining: 1.32s
509:	learn: 0.9635918	test: 0.7974181	best: 0.8301887 (74)	total: 1.37s	remaining: 1.31s
510:	learn: 0.9639419	test: 0.7974181	best: 0.8301887 (74)	total: 1.37s	remaining: 1.31s
511:	learn: 0.9637668	test: 0.7974181	best: 0.8301887 (74)	total: 1.37s	remaining: 1.31s
512:	learn: 0.9641169	test: 0.7974181	best: 0.8301887 (74)	total: 1.38s	remaining: 1.31s
513:	learn: 0.9641169	test: 0.7964250	best: 0.8301887 (74)	total: 1.38s	remaining: 1.3s
514:	learn: 0.9639419	test: 0.7964250	best: 0.8301887 (74)	total: 1.38s	remaining: 1.3s
515:	learn: 0.9641169	test: 0.7964250	best: 0.8301887 (74)	total: 1.39s	remaining: 1.3s
516:	learn: 0.9639419	test: 0.7964250	best: 0.8301887 (74)	total: 1.39s	remaining: 1.3s
517:	learn: 0.9639419	test: 0.7974181	best: 0.8301887 (74)	total: 1.39s	remaining: 1.29s
518:	learn: 0.9642920	test: 0.7994042	best: 0.8301887 (74)	total: 1.39s	remaining: 1.29s
519:	learn: 0.9641169	test: 0.7984111	best: 0.8301887 (74)	total: 1.4s	remaining: 1.29s
520:	learn: 0.9639419	test: 0.7984111	best: 0.8301887 (74)	total: 1.4s	remaining: 1.29s
521:	learn: 0.9642920	test: 0.7984111	best: 0.8301887 (74)	total: 1.4s	remaining: 1.28s
522:	learn: 0.9641169	test: 0.7984111	best: 0.8301887 (74)	total: 1.41s	remaining: 1.28s
523:	learn: 0.9641169	test: 0.7974181	best: 0.8301887 (74)	total: 1.41s	remaining: 1.28s
524:	learn: 0.9642920	test: 0.7974181	best: 0.8301887 (74)	total: 1.41s	remaining: 1.28s
525:	learn: 0.9644670	test: 0.7974181	best: 0.8301887 (74)	total: 1.41s	remaining: 1.27s
526:	learn: 0.9644670	test: 0.7974181	best: 0.8301887 (74)	total: 1.42s	remaining: 1.27s
527:	learn: 0.9646420	test: 0.7974181	best: 0.8301887 (74)	total: 1.42s	remaining: 1.27s
528:	learn: 0.9642920	test: 0.7974181	best: 0.8301887 (74)	total: 1.42s	remaining: 1.26s
529:	learn: 0.9642920	test: 0.7974181	best: 0.8301887 (74)	total: 1.42s	remaining: 1.26s
530:	learn: 0.9642920	test: 0.7974181	best: 0.8301887 (74)	total: 1.43s	remaining: 1.26s
531:	learn: 0.9642920	test: 0.7974181	best: 0.8301887 (74)	total: 1.43s	remaining: 1.26s
532:	learn: 0.9644670	test: 0.7974181	best: 0.8301887 (74)	total: 1.43s	remaining: 1.25s
533:	learn: 0.9646420	test: 0.7974181	best: 0.8301887 (74)	total: 1.43s	remaining: 1.25s
534:	learn: 0.9646420	test: 0.7974181	best: 0.8301887 (74)	total: 1.44s	remaining: 1.25s
535:	learn: 0.9655172	test: 0.7974181	best: 0.8301887 (74)	total: 1.44s	remaining: 1.25s
536:	learn: 0.9655172	test: 0.7974181	best: 0.8301887 (74)	total: 1.44s	remaining: 1.24s
537:	learn: 0.9655172	test: 0.7974181	best: 0.8301887 (74)	total: 1.44s	remaining: 1.24s
538:	learn: 0.9655172	test: 0.7974181	best: 0.8301887 (74)	total: 1.45s	remaining: 1.24s
539:	learn: 0.9656923	test: 0.7974181	best: 0.8301887 (74)	total: 1.45s	remaining: 1.24s
540:	learn: 0.9656923	test: 0.7974181	best: 0.8301887 (74)	total: 1.45s	remaining: 1.23s
541:	learn: 0.9656923	test: 0.7974181	best: 0.8301887 (74)	total: 1.46s	remaining: 1.23s
542:	learn: 0.9656923	test: 0.7974181	best: 0.8301887 (74)	total: 1.46s	remaining: 1.23s
543:	learn: 0.9656923	test: 0.7974181	best: 0.8301887 (74)	total: 1.46s	remaining: 1.23s
544:	learn: 0.9656923	test: 0.7974181	best: 0.8301887 (74)	total: 1.47s	remaining: 1.22s
545:	learn: 0.9658673	test: 0.7984111	best: 0.8301887 (74)	total: 1.47s	remaining: 1.22s
546:	learn: 0.9658673	test: 0.7984111	best: 0.8301887 (74)	total: 1.47s	remaining: 1.22s
547:	learn: 0.9658673	test: 0.7984111	best: 0.8301887 (74)	total: 1.47s	remaining: 1.22s
548:	learn: 0.9658673	test: 0.7984111	best: 0.8301887 (74)	total: 1.48s	remaining: 1.21s
549:	learn: 0.9670926	test: 0.7984111	best: 0.8301887 (74)	total: 1.48s	remaining: 1.21s
550:	learn: 0.9669176	test: 0.7984111	best: 0.8301887 (74)	total: 1.48s	remaining: 1.21s
551:	learn: 0.9670926	test: 0.7984111	best: 0.8301887 (74)	total: 1.48s	remaining: 1.21s
552:	learn: 0.9669176	test: 0.7984111	best: 0.8301887 (74)	total: 1.49s	remaining: 1.2s
553:	learn: 0.9669176	test: 0.7984111	best: 0.8301887 (74)	total: 1.49s	remaining: 1.2s
554:	learn: 0.9670926	test: 0.7984111	best: 0.8301887 (74)	total: 1.49s	remaining: 1.2s
555:	learn: 0.9669176	test: 0.7984111	best: 0.8301887 (74)	total: 1.5s	remaining: 1.19s
556:	learn: 0.9670926	test: 0.7984111	best: 0.8301887 (74)	total: 1.5s	remaining: 1.19s
557:	learn: 0.9672676	test: 0.7984111	best: 0.8301887 (74)	total: 1.5s	remaining: 1.19s
558:	learn: 0.9681428	test: 0.7994042	best: 0.8301887 (74)	total: 1.5s	remaining: 1.19s
559:	learn: 0.9681428	test: 0.7994042	best: 0.8301887 (74)	total: 1.51s	remaining: 1.18s
560:	learn: 0.9681428	test: 0.7984111	best: 0.8301887 (74)	total: 1.51s	remaining: 1.18s
561:	learn: 0.9681428	test: 0.7994042	best: 0.8301887 (74)	total: 1.51s	remaining: 1.18s
562:	learn: 0.9679678	test: 0.7934459	best: 0.8301887 (74)	total: 1.51s	remaining: 1.18s
563:	learn: 0.9684929	test: 0.7924528	best: 0.8301887 (74)	total: 1.52s	remaining: 1.17s
564:	learn: 0.9684929	test: 0.7924528	best: 0.8301887 (74)	total: 1.52s	remaining: 1.17s
565:	learn: 0.9684929	test: 0.7934459	best: 0.8301887 (74)	total: 1.52s	remaining: 1.17s
566:	learn: 0.9676177	test: 0.7934459	best: 0.8301887 (74)	total: 1.52s	remaining: 1.17s
567:	learn: 0.9686680	test: 0.7934459	best: 0.8301887 (74)	total: 1.53s	remaining: 1.16s
568:	learn: 0.9686680	test: 0.7934459	best: 0.8301887 (74)	total: 1.53s	remaining: 1.16s
569:	learn: 0.9688430	test: 0.7934459	best: 0.8301887 (74)	total: 1.53s	remaining: 1.16s
570:	learn: 0.9688430	test: 0.7934459	best: 0.8301887 (74)	total: 1.54s	remaining: 1.15s
571:	learn: 0.9688430	test: 0.7934459	best: 0.8301887 (74)	total: 1.54s	remaining: 1.15s
572:	learn: 0.9688430	test: 0.7934459	best: 0.8301887 (74)	total: 1.54s	remaining: 1.15s
573:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.54s	remaining: 1.15s
574:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.55s	remaining: 1.14s
575:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.55s	remaining: 1.14s
576:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.55s	remaining: 1.14s
577:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.55s	remaining: 1.14s
578:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.56s	remaining: 1.13s
579:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.56s	remaining: 1.13s
580:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.56s	remaining: 1.13s
581:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.57s	remaining: 1.12s
582:	learn: 0.9688430	test: 0.7944389	best: 0.8301887 (74)	total: 1.57s	remaining: 1.12s
583:	learn: 0.9688430	test: 0.7954320	best: 0.8301887 (74)	total: 1.57s	remaining: 1.12s
584:	learn: 0.9688430	test: 0.7954320	best: 0.8301887 (74)	total: 1.57s	remaining: 1.12s
585:	learn: 0.9690180	test: 0.7954320	best: 0.8301887 (74)	total: 1.58s	remaining: 1.11s
586:	learn: 0.9688430	test: 0.7954320	best: 0.8301887 (74)	total: 1.58s	remaining: 1.11s
587:	learn: 0.9688430	test: 0.7954320	best: 0.8301887 (74)	total: 1.58s	remaining: 1.11s
588:	learn: 0.9688430	test: 0.7954320	best: 0.8301887 (74)	total: 1.58s	remaining: 1.11s
589:	learn: 0.9688430	test: 0.7954320	best: 0.8301887 (74)	total: 1.59s	remaining: 1.1s
590:	learn: 0.9688430	test: 0.7954320	best: 0.8301887 (74)	total: 1.59s	remaining: 1.1s
591:	learn: 0.9690180	test: 0.7954320	best: 0.8301887 (74)	total: 1.59s	remaining: 1.1s
592:	learn: 0.9691931	test: 0.7954320	best: 0.8301887 (74)	total: 1.6s	remaining: 1.1s
593:	learn: 0.9691931	test: 0.7954320	best: 0.8301887 (74)	total: 1.6s	remaining: 1.09s
594:	learn: 0.9691931	test: 0.7954320	best: 0.8301887 (74)	total: 1.6s	remaining: 1.09s
595:	learn: 0.9691931	test: 0.7954320	best: 0.8301887 (74)	total: 1.6s	remaining: 1.09s
596:	learn: 0.9691931	test: 0.7954320	best: 0.8301887 (74)	total: 1.61s	remaining: 1.08s
597:	learn: 0.9700683	test: 0.7954320	best: 0.8301887 (74)	total: 1.61s	remaining: 1.08s
598:	learn: 0.9693681	test: 0.7894737	best: 0.8301887 (74)	total: 1.61s	remaining: 1.08s
599:	learn: 0.9702433	test: 0.7904667	best: 0.8301887 (74)	total: 1.62s	remaining: 1.08s
600:	learn: 0.9702433	test: 0.7904667	best: 0.8301887 (74)	total: 1.62s	remaining: 1.07s
601:	learn: 0.9702433	test: 0.7904667	best: 0.8301887 (74)	total: 1.62s	remaining: 1.07s
602:	learn: 0.9704183	test: 0.7904667	best: 0.8301887 (74)	total: 1.63s	remaining: 1.07s
603:	learn: 0.9700683	test: 0.7904667	best: 0.8301887 (74)	total: 1.63s	remaining: 1.07s
604:	learn: 0.9700683	test: 0.7904667	best: 0.8301887 (74)	total: 1.63s	remaining: 1.06s
605:	learn: 0.9707684	test: 0.7894737	best: 0.8301887 (74)	total: 1.63s	remaining: 1.06s
606:	learn: 0.9705934	test: 0.7894737	best: 0.8301887 (74)	total: 1.64s	remaining: 1.06s
607:	learn: 0.9714686	test: 0.7894737	best: 0.8301887 (74)	total: 1.64s	remaining: 1.06s
608:	learn: 0.9714686	test: 0.7894737	best: 0.8301887 (74)	total: 1.64s	remaining: 1.05s
609:	learn: 0.9712935	test: 0.7894737	best: 0.8301887 (74)	total: 1.64s	remaining: 1.05s
610:	learn: 0.9716436	test: 0.7894737	best: 0.8301887 (74)	total: 1.65s	remaining: 1.05s
611:	learn: 0.9716436	test: 0.7894737	best: 0.8301887 (74)	total: 1.65s	remaining: 1.04s
612:	learn: 0.9718187	test: 0.7894737	best: 0.8301887 (74)	total: 1.65s	remaining: 1.04s
613:	learn: 0.9712935	test: 0.7894737	best: 0.8301887 (74)	total: 1.66s	remaining: 1.04s
614:	learn: 0.9714686	test: 0.7894737	best: 0.8301887 (74)	total: 1.66s	remaining: 1.04s
615:	learn: 0.9714686	test: 0.7845084	best: 0.8301887 (74)	total: 1.66s	remaining: 1.03s
616:	learn: 0.9714686	test: 0.7894737	best: 0.8301887 (74)	total: 1.66s	remaining: 1.03s
617:	learn: 0.9714686	test: 0.7894737	best: 0.8301887 (74)	total: 1.67s	remaining: 1.03s
618:	learn: 0.9718187	test: 0.7845084	best: 0.8301887 (74)	total: 1.67s	remaining: 1.03s
619:	learn: 0.9716436	test: 0.7894737	best: 0.8301887 (74)	total: 1.67s	remaining: 1.02s
620:	learn: 0.9716436	test: 0.7894737	best: 0.8301887 (74)	total: 1.67s	remaining: 1.02s
621:	learn: 0.9716436	test: 0.7894737	best: 0.8301887 (74)	total: 1.68s	remaining: 1.02s
622:	learn: 0.9716436	test: 0.7894737	best: 0.8301887 (74)	total: 1.68s	remaining: 1.02s
623:	learn: 0.9716436	test: 0.7894737	best: 0.8301887 (74)	total: 1.68s	remaining: 1.01s
624:	learn: 0.9716436	test: 0.7894737	best: 0.8301887 (74)	total: 1.69s	remaining: 1.01s
625:	learn: 0.9718187	test: 0.7894737	best: 0.8301887 (74)	total: 1.69s	remaining: 1.01s
626:	learn: 0.9719937	test: 0.7894737	best: 0.8301887 (74)	total: 1.69s	remaining: 1s
627:	learn: 0.9719937	test: 0.7894737	best: 0.8301887 (74)	total: 1.69s	remaining: 1s
628:	learn: 0.9719937	test: 0.7894737	best: 0.8301887 (74)	total: 1.7s	remaining: 1s
629:	learn: 0.9721687	test: 0.7894737	best: 0.8301887 (74)	total: 1.7s	remaining: 998ms
630:	learn: 0.9721687	test: 0.7894737	best: 0.8301887 (74)	total: 1.7s	remaining: 996ms
631:	learn: 0.9721687	test: 0.7894737	best: 0.8301887 (74)	total: 1.71s	remaining: 993ms
632:	learn: 0.9721687	test: 0.7894737	best: 0.8301887 (74)	total: 1.71s	remaining: 990ms
633:	learn: 0.9721687	test: 0.7894737	best: 0.8301887 (74)	total: 1.71s	remaining: 988ms
634:	learn: 0.9721687	test: 0.7894737	best: 0.8301887 (74)	total: 1.71s	remaining: 985ms
635:	learn: 0.9721687	test: 0.7894737	best: 0.8301887 (74)	total: 1.72s	remaining: 982ms
636:	learn: 0.9721687	test: 0.7845084	best: 0.8301887 (74)	total: 1.72s	remaining: 979ms
637:	learn: 0.9721687	test: 0.7845084	best: 0.8301887 (74)	total: 1.72s	remaining: 977ms
638:	learn: 0.9723438	test: 0.7795432	best: 0.8301887 (74)	total: 1.72s	remaining: 974ms
639:	learn: 0.9723438	test: 0.7795432	best: 0.8301887 (74)	total: 1.73s	remaining: 971ms
640:	learn: 0.9723438	test: 0.7795432	best: 0.8301887 (74)	total: 1.73s	remaining: 969ms
641:	learn: 0.9723438	test: 0.7795432	best: 0.8301887 (74)	total: 1.73s	remaining: 966ms
642:	learn: 0.9723438	test: 0.7795432	best: 0.8301887 (74)	total: 1.73s	remaining: 963ms
643:	learn: 0.9723438	test: 0.7795432	best: 0.8301887 (74)	total: 1.74s	remaining: 960ms
644:	learn: 0.9725188	test: 0.7795432	best: 0.8301887 (74)	total: 1.74s	remaining: 958ms
645:	learn: 0.9723438	test: 0.7795432	best: 0.8301887 (74)	total: 1.74s	remaining: 955ms
646:	learn: 0.9723438	test: 0.7785501	best: 0.8301887 (74)	total: 1.75s	remaining: 952ms
647:	learn: 0.9725188	test: 0.7785501	best: 0.8301887 (74)	total: 1.75s	remaining: 950ms
648:	learn: 0.9725188	test: 0.7785501	best: 0.8301887 (74)	total: 1.75s	remaining: 947ms
649:	learn: 0.9726939	test: 0.7785501	best: 0.8301887 (74)	total: 1.75s	remaining: 944ms
650:	learn: 0.9726939	test: 0.7775571	best: 0.8301887 (74)	total: 1.76s	remaining: 941ms
651:	learn: 0.9728689	test: 0.7775571	best: 0.8301887 (74)	total: 1.76s	remaining: 939ms
652:	learn: 0.9730439	test: 0.7775571	best: 0.8301887 (74)	total: 1.76s	remaining: 936ms
653:	learn: 0.9730439	test: 0.7775571	best: 0.8301887 (74)	total: 1.76s	remaining: 934ms
654:	learn: 0.9730439	test: 0.7775571	best: 0.8301887 (74)	total: 1.77s	remaining: 931ms
655:	learn: 0.9733940	test: 0.7775571	best: 0.8301887 (74)	total: 1.77s	remaining: 928ms
656:	learn: 0.9733940	test: 0.7775571	best: 0.8301887 (74)	total: 1.77s	remaining: 926ms
657:	learn: 0.9733940	test: 0.7775571	best: 0.8301887 (74)	total: 1.78s	remaining: 923ms
658:	learn: 0.9735691	test: 0.7785501	best: 0.8301887 (74)	total: 1.78s	remaining: 921ms
659:	learn: 0.9732190	test: 0.7775571	best: 0.8301887 (74)	total: 1.78s	remaining: 918ms
660:	learn: 0.9732190	test: 0.7785501	best: 0.8301887 (74)	total: 1.79s	remaining: 916ms
661:	learn: 0.9733940	test: 0.7785501	best: 0.8301887 (74)	total: 1.79s	remaining: 913ms
662:	learn: 0.9737441	test: 0.7785501	best: 0.8301887 (74)	total: 1.79s	remaining: 911ms
663:	learn: 0.9739191	test: 0.7775571	best: 0.8301887 (74)	total: 1.79s	remaining: 908ms
664:	learn: 0.9740942	test: 0.7785501	best: 0.8301887 (74)	total: 1.8s	remaining: 905ms
665:	learn: 0.9744442	test: 0.7795432	best: 0.8301887 (74)	total: 1.8s	remaining: 903ms
666:	learn: 0.9744442	test: 0.7795432	best: 0.8301887 (74)	total: 1.8s	remaining: 900ms
667:	learn: 0.9742692	test: 0.7795432	best: 0.8301887 (74)	total: 1.8s	remaining: 897ms
668:	learn: 0.9742692	test: 0.7795432	best: 0.8301887 (74)	total: 1.81s	remaining: 895ms
669:	learn: 0.9742692	test: 0.7735849	best: 0.8301887 (74)	total: 1.81s	remaining: 892ms
670:	learn: 0.9742692	test: 0.7745780	best: 0.8301887 (74)	total: 1.81s	remaining: 889ms
671:	learn: 0.9740942	test: 0.7735849	best: 0.8301887 (74)	total: 1.82s	remaining: 887ms
672:	learn: 0.9740942	test: 0.7795432	best: 0.8301887 (74)	total: 1.82s	remaining: 884ms
673:	learn: 0.9740942	test: 0.7745780	best: 0.8301887 (74)	total: 1.82s	remaining: 882ms
674:	learn: 0.9742692	test: 0.7745780	best: 0.8301887 (74)	total: 1.83s	remaining: 879ms
675:	learn: 0.9746193	test: 0.7745780	best: 0.8301887 (74)	total: 1.83s	remaining: 877ms
676:	learn: 0.9746193	test: 0.7745780	best: 0.8301887 (74)	total: 1.83s	remaining: 874ms
677:	learn: 0.9746193	test: 0.7745780	best: 0.8301887 (74)	total: 1.83s	remaining: 871ms
678:	learn: 0.9747943	test: 0.7745780	best: 0.8301887 (74)	total: 1.84s	remaining: 868ms
679:	learn: 0.9760196	test: 0.7735849	best: 0.8301887 (74)	total: 1.84s	remaining: 866ms
680:	learn: 0.9761946	test: 0.7745780	best: 0.8301887 (74)	total: 1.84s	remaining: 863ms
681:	learn: 0.9761946	test: 0.7745780	best: 0.8301887 (74)	total: 1.84s	remaining: 860ms
682:	learn: 0.9761946	test: 0.7745780	best: 0.8301887 (74)	total: 1.85s	remaining: 857ms
683:	learn: 0.9761946	test: 0.7755710	best: 0.8301887 (74)	total: 1.85s	remaining: 855ms
684:	learn: 0.9763697	test: 0.7745780	best: 0.8301887 (74)	total: 1.85s	remaining: 852ms
685:	learn: 0.9760196	test: 0.7755710	best: 0.8301887 (74)	total: 1.85s	remaining: 849ms
686:	learn: 0.9770698	test: 0.7755710	best: 0.8301887 (74)	total: 1.86s	remaining: 847ms
687:	learn: 0.9763697	test: 0.7755710	best: 0.8301887 (74)	total: 1.86s	remaining: 844ms
688:	learn: 0.9782951	test: 0.7755710	best: 0.8301887 (74)	total: 1.86s	remaining: 842ms
689:	learn: 0.9772449	test: 0.7755710	best: 0.8301887 (74)	total: 1.87s	remaining: 839ms
690:	learn: 0.9774199	test: 0.7755710	best: 0.8301887 (74)	total: 1.87s	remaining: 837ms
691:	learn: 0.9774199	test: 0.7755710	best: 0.8301887 (74)	total: 1.87s	remaining: 834ms
692:	learn: 0.9774199	test: 0.7755710	best: 0.8301887 (74)	total: 1.88s	remaining: 831ms
693:	learn: 0.9774199	test: 0.7755710	best: 0.8301887 (74)	total: 1.88s	remaining: 829ms
694:	learn: 0.9765447	test: 0.7755710	best: 0.8301887 (74)	total: 1.88s	remaining: 826ms
695:	learn: 0.9767198	test: 0.7755710	best: 0.8301887 (74)	total: 1.88s	remaining: 823ms
696:	learn: 0.9775950	test: 0.7755710	best: 0.8301887 (74)	total: 1.89s	remaining: 820ms
697:	learn: 0.9775950	test: 0.7755710	best: 0.8301887 (74)	total: 1.89s	remaining: 818ms
698:	learn: 0.9777700	test: 0.7755710	best: 0.8301887 (74)	total: 1.89s	remaining: 815ms
699:	learn: 0.9777700	test: 0.7755710	best: 0.8301887 (74)	total: 1.9s	remaining: 812ms
700:	learn: 0.9777700	test: 0.7755710	best: 0.8301887 (74)	total: 1.9s	remaining: 810ms
701:	learn: 0.9777700	test: 0.7755710	best: 0.8301887 (74)	total: 1.9s	remaining: 807ms
702:	learn: 0.9779450	test: 0.7755710	best: 0.8301887 (74)	total: 1.9s	remaining: 804ms
703:	learn: 0.9788202	test: 0.7755710	best: 0.8301887 (74)	total: 1.91s	remaining: 802ms
704:	learn: 0.9779450	test: 0.7755710	best: 0.8301887 (74)	total: 1.91s	remaining: 799ms
705:	learn: 0.9788202	test: 0.7755710	best: 0.8301887 (74)	total: 1.91s	remaining: 796ms
706:	learn: 0.9786452	test: 0.7755710	best: 0.8301887 (74)	total: 1.92s	remaining: 794ms
707:	learn: 0.9786452	test: 0.7755710	best: 0.8301887 (74)	total: 1.92s	remaining: 791ms
708:	learn: 0.9786452	test: 0.7755710	best: 0.8301887 (74)	total: 1.92s	remaining: 788ms
709:	learn: 0.9786452	test: 0.7706058	best: 0.8301887 (74)	total: 1.92s	remaining: 785ms
710:	learn: 0.9786452	test: 0.7706058	best: 0.8301887 (74)	total: 1.93s	remaining: 783ms
711:	learn: 0.9786452	test: 0.7706058	best: 0.8301887 (74)	total: 1.93s	remaining: 780ms
712:	learn: 0.9782951	test: 0.7706058	best: 0.8301887 (74)	total: 1.93s	remaining: 777ms
713:	learn: 0.9782951	test: 0.7706058	best: 0.8301887 (74)	total: 1.93s	remaining: 774ms
714:	learn: 0.9788202	test: 0.7706058	best: 0.8301887 (74)	total: 1.94s	remaining: 772ms
715:	learn: 0.9786452	test: 0.7706058	best: 0.8301887 (74)	total: 1.94s	remaining: 769ms
716:	learn: 0.9786452	test: 0.7706058	best: 0.8301887 (74)	total: 1.94s	remaining: 766ms
717:	learn: 0.9786452	test: 0.7755710	best: 0.8301887 (74)	total: 1.94s	remaining: 763ms
718:	learn: 0.9786452	test: 0.7755710	best: 0.8301887 (74)	total: 1.95s	remaining: 760ms
719:	learn: 0.9786452	test: 0.7755710	best: 0.8301887 (74)	total: 1.95s	remaining: 758ms
720:	learn: 0.9786452	test: 0.7755710	best: 0.8301887 (74)	total: 1.95s	remaining: 755ms
721:	learn: 0.9788202	test: 0.7755710	best: 0.8301887 (74)	total: 1.95s	remaining: 752ms
722:	learn: 0.9788202	test: 0.7745780	best: 0.8301887 (74)	total: 1.96s	remaining: 749ms
723:	learn: 0.9789953	test: 0.7745780	best: 0.8301887 (74)	total: 1.96s	remaining: 746ms
724:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.96s	remaining: 744ms
725:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.96s	remaining: 741ms
726:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.97s	remaining: 739ms
727:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.97s	remaining: 736ms
728:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.97s	remaining: 733ms
729:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.97s	remaining: 730ms
730:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.98s	remaining: 728ms
731:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.98s	remaining: 725ms
732:	learn: 0.9791703	test: 0.7755710	best: 0.8301887 (74)	total: 1.98s	remaining: 722ms
733:	learn: 0.9791703	test: 0.7755710	best: 0.8301887 (74)	total: 1.99s	remaining: 720ms
734:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.99s	remaining: 717ms
735:	learn: 0.9789953	test: 0.7755710	best: 0.8301887 (74)	total: 1.99s	remaining: 715ms
736:	learn: 0.9791703	test: 0.7755710	best: 0.8301887 (74)	total: 2s	remaining: 712ms
737:	learn: 0.9791703	test: 0.7755710	best: 0.8301887 (74)	total: 2s	remaining: 710ms
738:	learn: 0.9791703	test: 0.7755710	best: 0.8301887 (74)	total: 2s	remaining: 707ms
739:	learn: 0.9793454	test: 0.7755710	best: 0.8301887 (74)	total: 2s	remaining: 704ms
740:	learn: 0.9791703	test: 0.7755710	best: 0.8301887 (74)	total: 2.01s	remaining: 702ms
741:	learn: 0.9791703	test: 0.7755710	best: 0.8301887 (74)	total: 2.01s	remaining: 699ms
742:	learn: 0.9791703	test: 0.7755710	best: 0.8301887 (74)	total: 2.01s	remaining: 696ms
743:	learn: 0.9793454	test: 0.7755710	best: 0.8301887 (74)	total: 2.01s	remaining: 693ms
744:	learn: 0.9793454	test: 0.7755710	best: 0.8301887 (74)	total: 2.02s	remaining: 690ms
745:	learn: 0.9793454	test: 0.7755710	best: 0.8301887 (74)	total: 2.02s	remaining: 688ms
746:	learn: 0.9793454	test: 0.7755710	best: 0.8301887 (74)	total: 2.02s	remaining: 685ms
747:	learn: 0.9793454	test: 0.7755710	best: 0.8301887 (74)	total: 2.02s	remaining: 682ms
748:	learn: 0.9793454	test: 0.7755710	best: 0.8301887 (74)	total: 2.03s	remaining: 679ms
749:	learn: 0.9793454	test: 0.7745780	best: 0.8301887 (74)	total: 2.03s	remaining: 677ms
750:	learn: 0.9793454	test: 0.7745780	best: 0.8301887 (74)	total: 2.03s	remaining: 674ms
751:	learn: 0.9793454	test: 0.7696127	best: 0.8301887 (74)	total: 2.03s	remaining: 671ms
752:	learn: 0.9793454	test: 0.7696127	best: 0.8301887 (74)	total: 2.04s	remaining: 668ms
753:	learn: 0.9796954	test: 0.7696127	best: 0.8301887 (74)	total: 2.04s	remaining: 665ms
754:	learn: 0.9791703	test: 0.7706058	best: 0.8301887 (74)	total: 2.04s	remaining: 663ms
755:	learn: 0.9793454	test: 0.7745780	best: 0.8301887 (74)	total: 2.04s	remaining: 660ms
756:	learn: 0.9793454	test: 0.7745780	best: 0.8301887 (74)	total: 2.05s	remaining: 657ms
757:	learn: 0.9796954	test: 0.7745780	best: 0.8301887 (74)	total: 2.05s	remaining: 654ms
758:	learn: 0.9798705	test: 0.7745780	best: 0.8301887 (74)	total: 2.05s	remaining: 652ms
759:	learn: 0.9800455	test: 0.7745780	best: 0.8301887 (74)	total: 2.06s	remaining: 649ms
760:	learn: 0.9800455	test: 0.7745780	best: 0.8301887 (74)	total: 2.06s	remaining: 646ms
761:	learn: 0.9802205	test: 0.7745780	best: 0.8301887 (74)	total: 2.06s	remaining: 643ms
762:	learn: 0.9802205	test: 0.7745780	best: 0.8301887 (74)	total: 2.06s	remaining: 641ms
763:	learn: 0.9798705	test: 0.7745780	best: 0.8301887 (74)	total: 2.06s	remaining: 638ms
764:	learn: 0.9800455	test: 0.7745780	best: 0.8301887 (74)	total: 2.07s	remaining: 635ms
765:	learn: 0.9803956	test: 0.7745780	best: 0.8301887 (74)	total: 2.07s	remaining: 633ms
766:	learn: 0.9814458	test: 0.7745780	best: 0.8301887 (74)	total: 2.07s	remaining: 630ms
767:	learn: 0.9814458	test: 0.7745780	best: 0.8301887 (74)	total: 2.08s	remaining: 627ms
768:	learn: 0.9814458	test: 0.7745780	best: 0.8301887 (74)	total: 2.08s	remaining: 624ms
769:	learn: 0.9814458	test: 0.7745780	best: 0.8301887 (74)	total: 2.08s	remaining: 622ms
770:	learn: 0.9816209	test: 0.7696127	best: 0.8301887 (74)	total: 2.08s	remaining: 619ms
771:	learn: 0.9816209	test: 0.7696127	best: 0.8301887 (74)	total: 2.09s	remaining: 616ms
772:	learn: 0.9816209	test: 0.7696127	best: 0.8301887 (74)	total: 2.09s	remaining: 613ms
773:	learn: 0.9816209	test: 0.7696127	best: 0.8301887 (74)	total: 2.09s	remaining: 611ms
774:	learn: 0.9816209	test: 0.7696127	best: 0.8301887 (74)	total: 2.09s	remaining: 608ms
775:	learn: 0.9816209	test: 0.7696127	best: 0.8301887 (74)	total: 2.1s	remaining: 605ms
776:	learn: 0.9817959	test: 0.7706058	best: 0.8301887 (74)	total: 2.1s	remaining: 603ms
777:	learn: 0.9816209	test: 0.7706058	best: 0.8301887 (74)	total: 2.1s	remaining: 600ms
778:	learn: 0.9816209	test: 0.7706058	best: 0.8301887 (74)	total: 2.1s	remaining: 597ms
779:	learn: 0.9819709	test: 0.7656405	best: 0.8301887 (74)	total: 2.11s	remaining: 594ms
780:	learn: 0.9819709	test: 0.7706058	best: 0.8301887 (74)	total: 2.11s	remaining: 592ms
781:	learn: 0.9817959	test: 0.7706058	best: 0.8301887 (74)	total: 2.11s	remaining: 589ms
782:	learn: 0.9817959	test: 0.7706058	best: 0.8301887 (74)	total: 2.12s	remaining: 586ms
783:	learn: 0.9817959	test: 0.7656405	best: 0.8301887 (74)	total: 2.12s	remaining: 583ms
784:	learn: 0.9819709	test: 0.7745780	best: 0.8301887 (74)	total: 2.12s	remaining: 581ms
785:	learn: 0.9817959	test: 0.7745780	best: 0.8301887 (74)	total: 2.12s	remaining: 578ms
786:	learn: 0.9819709	test: 0.7745780	best: 0.8301887 (74)	total: 2.13s	remaining: 575ms
787:	learn: 0.9819709	test: 0.7696127	best: 0.8301887 (74)	total: 2.13s	remaining: 573ms
788:	learn: 0.9821460	test: 0.7696127	best: 0.8301887 (74)	total: 2.13s	remaining: 570ms
789:	learn: 0.9817959	test: 0.7696127	best: 0.8301887 (74)	total: 2.13s	remaining: 567ms
790:	learn: 0.9817959	test: 0.7696127	best: 0.8301887 (74)	total: 2.14s	remaining: 565ms
791:	learn: 0.9817959	test: 0.7696127	best: 0.8301887 (74)	total: 2.14s	remaining: 562ms
792:	learn: 0.9819709	test: 0.7696127	best: 0.8301887 (74)	total: 2.14s	remaining: 559ms
793:	learn: 0.9817959	test: 0.7686197	best: 0.8301887 (74)	total: 2.14s	remaining: 556ms
794:	learn: 0.9819709	test: 0.7696127	best: 0.8301887 (74)	total: 2.15s	remaining: 554ms
795:	learn: 0.9819709	test: 0.7696127	best: 0.8301887 (74)	total: 2.15s	remaining: 551ms
796:	learn: 0.9819709	test: 0.7696127	best: 0.8301887 (74)	total: 2.15s	remaining: 548ms
797:	learn: 0.9821460	test: 0.7696127	best: 0.8301887 (74)	total: 2.15s	remaining: 545ms
798:	learn: 0.9830212	test: 0.7686197	best: 0.8301887 (74)	total: 2.16s	remaining: 542ms
799:	learn: 0.9828461	test: 0.7686197	best: 0.8301887 (74)	total: 2.16s	remaining: 540ms
800:	learn: 0.9828461	test: 0.7696127	best: 0.8301887 (74)	total: 2.16s	remaining: 537ms
801:	learn: 0.9830212	test: 0.7696127	best: 0.8301887 (74)	total: 2.16s	remaining: 534ms
802:	learn: 0.9831962	test: 0.7696127	best: 0.8301887 (74)	total: 2.17s	remaining: 531ms
803:	learn: 0.9831962	test: 0.7696127	best: 0.8301887 (74)	total: 2.17s	remaining: 529ms
804:	learn: 0.9833713	test: 0.7696127	best: 0.8301887 (74)	total: 2.17s	remaining: 526ms
805:	learn: 0.9833713	test: 0.7696127	best: 0.8301887 (74)	total: 2.17s	remaining: 523ms
806:	learn: 0.9833713	test: 0.7696127	best: 0.8301887 (74)	total: 2.18s	remaining: 521ms
807:	learn: 0.9833713	test: 0.7696127	best: 0.8301887 (74)	total: 2.18s	remaining: 518ms
808:	learn: 0.9831962	test: 0.7696127	best: 0.8301887 (74)	total: 2.18s	remaining: 515ms
809:	learn: 0.9833713	test: 0.7706058	best: 0.8301887 (74)	total: 2.19s	remaining: 513ms
810:	learn: 0.9833713	test: 0.7706058	best: 0.8301887 (74)	total: 2.19s	remaining: 510ms
811:	learn: 0.9833713	test: 0.7706058	best: 0.8301887 (74)	total: 2.19s	remaining: 508ms
812:	learn: 0.9831962	test: 0.7706058	best: 0.8301887 (74)	total: 2.2s	remaining: 505ms
813:	learn: 0.9833713	test: 0.7706058	best: 0.8301887 (74)	total: 2.2s	remaining: 503ms
814:	learn: 0.9833713	test: 0.7706058	best: 0.8301887 (74)	total: 2.2s	remaining: 500ms
815:	learn: 0.9833713	test: 0.7686197	best: 0.8301887 (74)	total: 2.21s	remaining: 497ms
816:	learn: 0.9833713	test: 0.7686197	best: 0.8301887 (74)	total: 2.21s	remaining: 495ms
817:	learn: 0.9833713	test: 0.7686197	best: 0.8301887 (74)	total: 2.21s	remaining: 492ms
818:	learn: 0.9833713	test: 0.7686197	best: 0.8301887 (74)	total: 2.21s	remaining: 489ms
819:	learn: 0.9833713	test: 0.7686197	best: 0.8301887 (74)	total: 2.21s	remaining: 486ms
820:	learn: 0.9833713	test: 0.7686197	best: 0.8301887 (74)	total: 2.22s	remaining: 483ms
821:	learn: 0.9833713	test: 0.7696127	best: 0.8301887 (74)	total: 2.22s	remaining: 481ms
822:	learn: 0.9833713	test: 0.7686197	best: 0.8301887 (74)	total: 2.22s	remaining: 478ms
823:	learn: 0.9835463	test: 0.7686197	best: 0.8301887 (74)	total: 2.23s	remaining: 475ms
824:	learn: 0.9835463	test: 0.7656405	best: 0.8301887 (74)	total: 2.23s	remaining: 473ms
825:	learn: 0.9835463	test: 0.7656405	best: 0.8301887 (74)	total: 2.23s	remaining: 470ms
826:	learn: 0.9835463	test: 0.7706058	best: 0.8301887 (74)	total: 2.23s	remaining: 467ms
827:	learn: 0.9835463	test: 0.7715988	best: 0.8301887 (74)	total: 2.24s	remaining: 465ms
828:	learn: 0.9835463	test: 0.7715988	best: 0.8301887 (74)	total: 2.24s	remaining: 462ms
829:	learn: 0.9833713	test: 0.7715988	best: 0.8301887 (74)	total: 2.24s	remaining: 459ms
830:	learn: 0.9833713	test: 0.7715988	best: 0.8301887 (74)	total: 2.24s	remaining: 456ms
831:	learn: 0.9833713	test: 0.7715988	best: 0.8301887 (74)	total: 2.25s	remaining: 454ms
832:	learn: 0.9835463	test: 0.7715988	best: 0.8301887 (74)	total: 2.25s	remaining: 451ms
833:	learn: 0.9835463	test: 0.7715988	best: 0.8301887 (74)	total: 2.25s	remaining: 448ms
834:	learn: 0.9835463	test: 0.7676266	best: 0.8301887 (74)	total: 2.25s	remaining: 445ms
835:	learn: 0.9835463	test: 0.7666336	best: 0.8301887 (74)	total: 2.26s	remaining: 443ms
836:	learn: 0.9835463	test: 0.7715988	best: 0.8301887 (74)	total: 2.26s	remaining: 440ms
837:	learn: 0.9835463	test: 0.7666336	best: 0.8301887 (74)	total: 2.26s	remaining: 437ms
838:	learn: 0.9833713	test: 0.7666336	best: 0.8301887 (74)	total: 2.26s	remaining: 435ms
839:	learn: 0.9833713	test: 0.7686197	best: 0.8301887 (74)	total: 2.27s	remaining: 432ms
840:	learn: 0.9833713	test: 0.7735849	best: 0.8301887 (74)	total: 2.27s	remaining: 429ms
841:	learn: 0.9833713	test: 0.7735849	best: 0.8301887 (74)	total: 2.27s	remaining: 426ms
842:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.27s	remaining: 424ms
843:	learn: 0.9833713	test: 0.7725919	best: 0.8301887 (74)	total: 2.28s	remaining: 421ms
844:	learn: 0.9833713	test: 0.7725919	best: 0.8301887 (74)	total: 2.28s	remaining: 418ms
845:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.28s	remaining: 415ms
846:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.28s	remaining: 413ms
847:	learn: 0.9833713	test: 0.7666336	best: 0.8301887 (74)	total: 2.29s	remaining: 410ms
848:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.29s	remaining: 407ms
849:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.29s	remaining: 405ms
850:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.29s	remaining: 402ms
851:	learn: 0.9833713	test: 0.7666336	best: 0.8301887 (74)	total: 2.3s	remaining: 399ms
852:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.3s	remaining: 396ms
853:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.3s	remaining: 394ms
854:	learn: 0.9835463	test: 0.7666336	best: 0.8301887 (74)	total: 2.31s	remaining: 391ms
855:	learn: 0.9835463	test: 0.7666336	best: 0.8301887 (74)	total: 2.31s	remaining: 388ms
856:	learn: 0.9835463	test: 0.7666336	best: 0.8301887 (74)	total: 2.31s	remaining: 386ms
857:	learn: 0.9833713	test: 0.7676266	best: 0.8301887 (74)	total: 2.31s	remaining: 383ms
858:	learn: 0.9835463	test: 0.7676266	best: 0.8301887 (74)	total: 2.31s	remaining: 380ms
859:	learn: 0.9837213	test: 0.7676266	best: 0.8301887 (74)	total: 2.32s	remaining: 377ms
860:	learn: 0.9835463	test: 0.7676266	best: 0.8301887 (74)	total: 2.32s	remaining: 375ms
861:	learn: 0.9837213	test: 0.7676266	best: 0.8301887 (74)	total: 2.32s	remaining: 372ms
862:	learn: 0.9835463	test: 0.7676266	best: 0.8301887 (74)	total: 2.33s	remaining: 369ms
863:	learn: 0.9835463	test: 0.7676266	best: 0.8301887 (74)	total: 2.33s	remaining: 367ms
864:	learn: 0.9838964	test: 0.7676266	best: 0.8301887 (74)	total: 2.33s	remaining: 364ms
865:	learn: 0.9840714	test: 0.7676266	best: 0.8301887 (74)	total: 2.33s	remaining: 361ms
866:	learn: 0.9840714	test: 0.7676266	best: 0.8301887 (74)	total: 2.34s	remaining: 359ms
867:	learn: 0.9840714	test: 0.7676266	best: 0.8301887 (74)	total: 2.34s	remaining: 356ms
868:	learn: 0.9840714	test: 0.7676266	best: 0.8301887 (74)	total: 2.34s	remaining: 353ms
869:	learn: 0.9840714	test: 0.7676266	best: 0.8301887 (74)	total: 2.34s	remaining: 350ms
870:	learn: 0.9840714	test: 0.7676266	best: 0.8301887 (74)	total: 2.35s	remaining: 348ms
871:	learn: 0.9840714	test: 0.7676266	best: 0.8301887 (74)	total: 2.35s	remaining: 345ms
872:	learn: 0.9840714	test: 0.7676266	best: 0.8301887 (74)	total: 2.35s	remaining: 342ms
873:	learn: 0.9842465	test: 0.7676266	best: 0.8301887 (74)	total: 2.35s	remaining: 340ms
874:	learn: 0.9842465	test: 0.7676266	best: 0.8301887 (74)	total: 2.36s	remaining: 337ms
875:	learn: 0.9842465	test: 0.7676266	best: 0.8301887 (74)	total: 2.36s	remaining: 334ms
876:	learn: 0.9844215	test: 0.7676266	best: 0.8301887 (74)	total: 2.36s	remaining: 332ms
877:	learn: 0.9842465	test: 0.7676266	best: 0.8301887 (74)	total: 2.37s	remaining: 329ms
878:	learn: 0.9838964	test: 0.7676266	best: 0.8301887 (74)	total: 2.37s	remaining: 326ms
879:	learn: 0.9838964	test: 0.7676266	best: 0.8301887 (74)	total: 2.37s	remaining: 324ms
880:	learn: 0.9838964	test: 0.7676266	best: 0.8301887 (74)	total: 2.38s	remaining: 321ms
881:	learn: 0.9838964	test: 0.7666336	best: 0.8301887 (74)	total: 2.38s	remaining: 318ms
882:	learn: 0.9838964	test: 0.7656405	best: 0.8301887 (74)	total: 2.38s	remaining: 315ms
883:	learn: 0.9838964	test: 0.7656405	best: 0.8301887 (74)	total: 2.38s	remaining: 313ms
884:	learn: 0.9838964	test: 0.7656405	best: 0.8301887 (74)	total: 2.39s	remaining: 310ms
885:	learn: 0.9838964	test: 0.7656405	best: 0.8301887 (74)	total: 2.39s	remaining: 307ms
886:	learn: 0.9840714	test: 0.7656405	best: 0.8301887 (74)	total: 2.39s	remaining: 305ms
887:	learn: 0.9840714	test: 0.7666336	best: 0.8301887 (74)	total: 2.39s	remaining: 302ms
888:	learn: 0.9840714	test: 0.7656405	best: 0.8301887 (74)	total: 2.4s	remaining: 299ms
889:	learn: 0.9842465	test: 0.7656405	best: 0.8301887 (74)	total: 2.4s	remaining: 297ms
890:	learn: 0.9844215	test: 0.7656405	best: 0.8301887 (74)	total: 2.4s	remaining: 294ms
891:	learn: 0.9844215	test: 0.7656405	best: 0.8301887 (74)	total: 2.41s	remaining: 292ms
892:	learn: 0.9844215	test: 0.7656405	best: 0.8301887 (74)	total: 2.41s	remaining: 289ms
893:	learn: 0.9844215	test: 0.7656405	best: 0.8301887 (74)	total: 2.41s	remaining: 286ms
894:	learn: 0.9844215	test: 0.7656405	best: 0.8301887 (74)	total: 2.42s	remaining: 284ms
895:	learn: 0.9842465	test: 0.7656405	best: 0.8301887 (74)	total: 2.42s	remaining: 281ms
896:	learn: 0.9842465	test: 0.7656405	best: 0.8301887 (74)	total: 2.42s	remaining: 278ms
897:	learn: 0.9844215	test: 0.7656405	best: 0.8301887 (74)	total: 2.43s	remaining: 276ms
898:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.43s	remaining: 273ms
899:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.43s	remaining: 270ms
900:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.44s	remaining: 268ms
901:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.44s	remaining: 265ms
902:	learn: 0.9847716	test: 0.7656405	best: 0.8301887 (74)	total: 2.44s	remaining: 262ms
903:	learn: 0.9847716	test: 0.7656405	best: 0.8301887 (74)	total: 2.44s	remaining: 260ms
904:	learn: 0.9847716	test: 0.7656405	best: 0.8301887 (74)	total: 2.45s	remaining: 257ms
905:	learn: 0.9847716	test: 0.7656405	best: 0.8301887 (74)	total: 2.45s	remaining: 254ms
906:	learn: 0.9847716	test: 0.7656405	best: 0.8301887 (74)	total: 2.45s	remaining: 251ms
907:	learn: 0.9847716	test: 0.7656405	best: 0.8301887 (74)	total: 2.46s	remaining: 249ms
908:	learn: 0.9847716	test: 0.7676266	best: 0.8301887 (74)	total: 2.46s	remaining: 246ms
909:	learn: 0.9847716	test: 0.7666336	best: 0.8301887 (74)	total: 2.46s	remaining: 243ms
910:	learn: 0.9849466	test: 0.7656405	best: 0.8301887 (74)	total: 2.46s	remaining: 241ms
911:	learn: 0.9849466	test: 0.7656405	best: 0.8301887 (74)	total: 2.47s	remaining: 238ms
912:	learn: 0.9851217	test: 0.7646475	best: 0.8301887 (74)	total: 2.47s	remaining: 235ms
913:	learn: 0.9847716	test: 0.7646475	best: 0.8301887 (74)	total: 2.47s	remaining: 233ms
914:	learn: 0.9847716	test: 0.7646475	best: 0.8301887 (74)	total: 2.48s	remaining: 230ms
915:	learn: 0.9847716	test: 0.7646475	best: 0.8301887 (74)	total: 2.48s	remaining: 227ms
916:	learn: 0.9847716	test: 0.7696127	best: 0.8301887 (74)	total: 2.48s	remaining: 225ms
917:	learn: 0.9849466	test: 0.7696127	best: 0.8301887 (74)	total: 2.48s	remaining: 222ms
918:	learn: 0.9849466	test: 0.7696127	best: 0.8301887 (74)	total: 2.49s	remaining: 219ms
919:	learn: 0.9849466	test: 0.7646475	best: 0.8301887 (74)	total: 2.49s	remaining: 216ms
920:	learn: 0.9847716	test: 0.7696127	best: 0.8301887 (74)	total: 2.49s	remaining: 214ms
921:	learn: 0.9845965	test: 0.7646475	best: 0.8301887 (74)	total: 2.49s	remaining: 211ms
922:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.5s	remaining: 208ms
923:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.5s	remaining: 206ms
924:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.5s	remaining: 203ms
925:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.5s	remaining: 200ms
926:	learn: 0.9845965	test: 0.7656405	best: 0.8301887 (74)	total: 2.51s	remaining: 197ms
927:	learn: 0.9847716	test: 0.7656405	best: 0.8301887 (74)	total: 2.51s	remaining: 195ms
928:	learn: 0.9849466	test: 0.7656405	best: 0.8301887 (74)	total: 2.51s	remaining: 192ms
929:	learn: 0.9851217	test: 0.7646475	best: 0.8301887 (74)	total: 2.52s	remaining: 189ms
930:	learn: 0.9851217	test: 0.7646475	best: 0.8301887 (74)	total: 2.52s	remaining: 187ms
931:	learn: 0.9851217	test: 0.7656405	best: 0.8301887 (74)	total: 2.52s	remaining: 184ms
932:	learn: 0.9854717	test: 0.7646475	best: 0.8301887 (74)	total: 2.52s	remaining: 181ms
933:	learn: 0.9852967	test: 0.7656405	best: 0.8301887 (74)	total: 2.53s	remaining: 179ms
934:	learn: 0.9851217	test: 0.7656405	best: 0.8301887 (74)	total: 2.53s	remaining: 176ms
935:	learn: 0.9852967	test: 0.7666336	best: 0.8301887 (74)	total: 2.53s	remaining: 173ms
936:	learn: 0.9852967	test: 0.7666336	best: 0.8301887 (74)	total: 2.54s	remaining: 170ms
937:	learn: 0.9852967	test: 0.7606753	best: 0.8301887 (74)	total: 2.54s	remaining: 168ms
938:	learn: 0.9852967	test: 0.7606753	best: 0.8301887 (74)	total: 2.54s	remaining: 165ms
939:	learn: 0.9854717	test: 0.7606753	best: 0.8301887 (74)	total: 2.54s	remaining: 162ms
940:	learn: 0.9856468	test: 0.7606753	best: 0.8301887 (74)	total: 2.55s	remaining: 160ms
941:	learn: 0.9858218	test: 0.7606753	best: 0.8301887 (74)	total: 2.55s	remaining: 157ms
942:	learn: 0.9859968	test: 0.7606753	best: 0.8301887 (74)	total: 2.55s	remaining: 154ms
943:	learn: 0.9859968	test: 0.7606753	best: 0.8301887 (74)	total: 2.55s	remaining: 152ms
944:	learn: 0.9858218	test: 0.7606753	best: 0.8301887 (74)	total: 2.56s	remaining: 149ms
945:	learn: 0.9861719	test: 0.7606753	best: 0.8301887 (74)	total: 2.56s	remaining: 146ms
946:	learn: 0.9863469	test: 0.7606753	best: 0.8301887 (74)	total: 2.56s	remaining: 143ms
947:	learn: 0.9865220	test: 0.7656405	best: 0.8301887 (74)	total: 2.56s	remaining: 141ms
948:	learn: 0.9865220	test: 0.7656405	best: 0.8301887 (74)	total: 2.57s	remaining: 138ms
949:	learn: 0.9866970	test: 0.7656405	best: 0.8301887 (74)	total: 2.57s	remaining: 135ms
950:	learn: 0.9866970	test: 0.7656405	best: 0.8301887 (74)	total: 2.58s	remaining: 133ms
951:	learn: 0.9866970	test: 0.7656405	best: 0.8301887 (74)	total: 2.58s	remaining: 130ms
952:	learn: 0.9866970	test: 0.7656405	best: 0.8301887 (74)	total: 2.58s	remaining: 127ms
953:	learn: 0.9866970	test: 0.7656405	best: 0.8301887 (74)	total: 2.58s	remaining: 125ms
954:	learn: 0.9866970	test: 0.7656405	best: 0.8301887 (74)	total: 2.59s	remaining: 122ms
955:	learn: 0.9865220	test: 0.7656405	best: 0.8301887 (74)	total: 2.59s	remaining: 119ms
956:	learn: 0.9863469	test: 0.7666336	best: 0.8301887 (74)	total: 2.59s	remaining: 117ms
957:	learn: 0.9865220	test: 0.7616683	best: 0.8301887 (74)	total: 2.6s	remaining: 114ms
958:	learn: 0.9868720	test: 0.7666336	best: 0.8301887 (74)	total: 2.6s	remaining: 111ms
959:	learn: 0.9868720	test: 0.7666336	best: 0.8301887 (74)	total: 2.6s	remaining: 108ms
960:	learn: 0.9866970	test: 0.7666336	best: 0.8301887 (74)	total: 2.6s	remaining: 106ms
961:	learn: 0.9866970	test: 0.7666336	best: 0.8301887 (74)	total: 2.61s	remaining: 103ms
962:	learn: 0.9870471	test: 0.7686197	best: 0.8301887 (74)	total: 2.61s	remaining: 100ms
963:	learn: 0.9870471	test: 0.7686197	best: 0.8301887 (74)	total: 2.61s	remaining: 97.6ms
964:	learn: 0.9872221	test: 0.7686197	best: 0.8301887 (74)	total: 2.62s	remaining: 94.9ms
965:	learn: 0.9872221	test: 0.7686197	best: 0.8301887 (74)	total: 2.62s	remaining: 92.2ms
966:	learn: 0.9873972	test: 0.7676266	best: 0.8301887 (74)	total: 2.62s	remaining: 89.5ms
967:	learn: 0.9873972	test: 0.7626614	best: 0.8301887 (74)	total: 2.62s	remaining: 86.8ms
968:	learn: 0.9877472	test: 0.7676266	best: 0.8301887 (74)	total: 2.63s	remaining: 84ms
969:	learn: 0.9877472	test: 0.7676266	best: 0.8301887 (74)	total: 2.63s	remaining: 81.3ms
970:	learn: 0.9879223	test: 0.7626614	best: 0.8301887 (74)	total: 2.63s	remaining: 78.6ms
971:	learn: 0.9879223	test: 0.7616683	best: 0.8301887 (74)	total: 2.63s	remaining: 75.9ms
972:	learn: 0.9879223	test: 0.7626614	best: 0.8301887 (74)	total: 2.64s	remaining: 73.2ms
973:	learn: 0.9879223	test: 0.7626614	best: 0.8301887 (74)	total: 2.64s	remaining: 70.5ms
974:	learn: 0.9879223	test: 0.7626614	best: 0.8301887 (74)	total: 2.64s	remaining: 67.8ms
975:	learn: 0.9879223	test: 0.7626614	best: 0.8301887 (74)	total: 2.65s	remaining: 65.1ms
976:	learn: 0.9879223	test: 0.7626614	best: 0.8301887 (74)	total: 2.65s	remaining: 62.4ms
977:	learn: 0.9879223	test: 0.7626614	best: 0.8301887 (74)	total: 2.65s	remaining: 59.7ms
978:	learn: 0.9879223	test: 0.7636544	best: 0.8301887 (74)	total: 2.65s	remaining: 57ms
979:	learn: 0.9877472	test: 0.7646475	best: 0.8301887 (74)	total: 2.66s	remaining: 54.3ms
980:	learn: 0.9879223	test: 0.7646475	best: 0.8301887 (74)	total: 2.66s	remaining: 51.5ms
981:	learn: 0.9879223	test: 0.7646475	best: 0.8301887 (74)	total: 2.66s	remaining: 48.8ms
982:	learn: 0.9880973	test: 0.7646475	best: 0.8301887 (74)	total: 2.67s	remaining: 46.1ms
983:	learn: 0.9880973	test: 0.7646475	best: 0.8301887 (74)	total: 2.67s	remaining: 43.4ms
984:	learn: 0.9882724	test: 0.7646475	best: 0.8301887 (74)	total: 2.67s	remaining: 40.7ms
985:	learn: 0.9882724	test: 0.7696127	best: 0.8301887 (74)	total: 2.68s	remaining: 38ms
986:	learn: 0.9882724	test: 0.7696127	best: 0.8301887 (74)	total: 2.68s	remaining: 35.3ms
987:	learn: 0.9882724	test: 0.7686197	best: 0.8301887 (74)	total: 2.68s	remaining: 32.6ms
988:	learn: 0.9882724	test: 0.7686197	best: 0.8301887 (74)	total: 2.69s	remaining: 29.9ms
989:	learn: 0.9887975	test: 0.7686197	best: 0.8301887 (74)	total: 2.69s	remaining: 27.2ms
990:	learn: 0.9887975	test: 0.7636544	best: 0.8301887 (74)	total: 2.69s	remaining: 24.4ms
991:	learn: 0.9887975	test: 0.7636544	best: 0.8301887 (74)	total: 2.69s	remaining: 21.7ms
992:	learn: 0.9887975	test: 0.7636544	best: 0.8301887 (74)	total: 2.7s	remaining: 19ms
993:	learn: 0.9887975	test: 0.7636544	best: 0.8301887 (74)	total: 2.7s	remaining: 16.3ms
994:	learn: 0.9886224	test: 0.7636544	best: 0.8301887 (74)	total: 2.7s	remaining: 13.6ms
995:	learn: 0.9887975	test: 0.7636544	best: 0.8301887 (74)	total: 2.71s	remaining: 10.9ms
996:	learn: 0.9887975	test: 0.7636544	best: 0.8301887 (74)	total: 2.71s	remaining: 8.15ms
997:	learn: 0.9887975	test: 0.7636544	best: 0.8301887 (74)	total: 2.71s	remaining: 5.43ms
998:	learn: 0.9891476	test: 0.7636544	best: 0.8301887 (74)	total: 2.71s	remaining: 2.72ms
999:	learn: 0.9891476	test: 0.7636544	best: 0.8301887 (74)	total: 2.72s	remaining: 0us

bestTest = 0.8301886792
bestIteration = 74

Shrink model to first 75 iterations.
0.8533333333333333
XGBoost Modeling¶
In [45]:
## Sine XGBoost does not care about categorical eatures on its own, thereore we have to convert categorical features into numerical
cat_cols = preprocess_data.select_dtypes(include=['object']).columns.tolist()
dummy_num_data = pd.get_dummies(preprocess_data[cat_cols])

num_cols = preprocess_data.select_dtypes(exclude=['object'])

dummy_train_data = pd.concat([dummy_num_data, num_cols],axis=1)
In [46]:
## Prepare a test data, which will work as unforseen data
test_data = dummy_train_data.sample(4500).reset_index(drop=True)
train_data = dummy_train_data.drop(test_data.index, axis=0).reset_index(drop=True)

## Segregate dependent variable from indepedent variables 
train_data_X = train_data.drop(['y'],axis=1)
train_data_Y = train_data.y

test_data_X = test_data.drop(['y'],axis=1)
test_data_Y = test_data.y
In [47]:
## XGBClassifier

## Using StandardScaler to standardize the features
sc = StandardScaler()
X = train_data_X
X = sc.fit_transform(X)
X = pd.DataFrame(X)

test_data = test_data_X
test_data = sc.fit_transform(test_data)
test_data = pd.DataFrame(test_data)

X_train, X_val, y_train, y_val = train_test_split(X,train_data_Y,train_size=.85,random_state=1234)
          
def train_and_predict(model,X_train,y_train,X_val,y_val,test_data):
    model.fit(X_train, y_train)
    y_pred = model.predict(X_val)
    print(1-mean_absolute_error(y_val, y_pred))
    return model.predict_proba(test_data)[:,1]


kf = KFold(n_splits=5, shuffle=True, random_state=3)
pred_arr = []
model = XGBClassifier(n_jobs=-1, random_state=42, n_estimators=500, scale_pos_weight = 5,)
for train_index, test_index in kf.split(X):    
    pred =train_and_predict(model,X.iloc[train_index],train_data_Y.iloc[train_index],X.iloc[test_index],train_data_Y.iloc[test_index],test_data)
    pred_arr.append(pred)

[18:56:38] WARNING: ../src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
0.8884794890690249
[18:56:45] WARNING: ../src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
0.8849177106362073
[18:56:52] WARNING: ../src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
0.8894620486366986
[18:57:00] WARNING: ../src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
0.8888479489069024
[18:57:08] WARNING: ../src/learner.cc:1115: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.
0.8937607467452714
In [48]:
final_y_pred = (pred_arr[0]+pred_arr[1]+pred_arr[2]+pred_arr[3]+pred_arr[4]+final_pred[:,1])/6
final_y_pred
Out[48]:
array([0.07555678, 0.05826938, 0.07859755, ..., 0.02231362, 0.43762432,
       0.37395117])
In [49]:
accuracy = 1-mean_absolute_error(test_data_Y, final_y_pred)
print('The accuracy of the above emsembled model is {0}'.format(round(accuracy,2)))

The accuracy of the above emsembled model is 0.88
In [ ]:
In [ ]:
In [ ]:
In [ ]: